Raincoast Applied Ecology - CANADIAN JOURNAL …...The Biology of Invasive Alien Plants in Canada....

CANADIAN JOURNAL OF

PLANTSCIENCE

REVUE CANADIENNE DE PHYTOTECHNIE

VOLUME 86 NO. 2 APRIL/AVRIL 2006

The Biology of Invasive Alien Plants in Canada. 4.Heracleum mantegazzianum Sommier & Levier

Nicholas A. Page1, Ronald E. Wall2, Stephen J. Darbyshire3, and Gerald A. Mulligan3

1Raincoast Applied Ecology, 102-1661 West 2nd Avenue, Vancouver, British Columbia, Canada, V6J 1H3 (e-mail: [email protected]); 29-454 Morison Avenue, Parksville, British Columbia, Canada, V9P 2M6;

3Agriculture and Agri-Food Canada, Central Experimental Farm, Saunders Building #49, Ottawa, Ontario,Canada K1A 0C6. Received 17 August 2005, accepted 20 December 2005.

Page, N. A., Wall, R. E., Darbyshire, S. J. and Mulligan, G. A. 2006. The Biology of Invasive Alien Plants in Canada. 4.Heracleum mantegazzianum Sommier & Levier. Can. J. Plant Sci. 86: 569–589. Heracleum mantegazzianum (giant hogweed)is an invasive alien plant of management concern in southern Canada where it has escaped from horticulture and established andspread in natural, ruderal, and agricultural ecosystems. It poses a threat to natural ecosystems and human health, and is also a weedin agricultural and urban areas. It is a member of the Carrot family (Apiaceae) and is closely related to the native speciesHeracleum maximum Bartram (cow-parsnip). It is a monocarpic perennial, which generally flowers in its 3rd or 4th year. Largesize, leaf shape, dark reddish pigments in patches on stems and petioles, and fruit characteristics readily distinguish H. man-tegazzianum from other plants in Canada. It is increasingly common in riparian areas, floodplains, and forest edges in or near urbanareas in southwestern British Columbia and southern Ontario. Based on herbarium specimens, H. mantegazzianum was firstrecorded in Ontario in 1949, British Columbia in 1964, Nova Scotia in 1980, Quebec in 1990, and New Brunswick in 2000. Thedevelopment of dense stands of H. mantegazzianum can also reduce the richness of native plants. Contact with H. mantegazzianumcan cause phytophotodermatitis, a serious skin inflammation caused by UV photo-activation of furanocoumarins present in the sap.Control methods include herbicide application, mechanical cutting, and animal grazing, but strategies to address seed dispersal andre-establishment from dormant seed must also be adopted. Widespread establishment in southern Canada suggests that eradicationis unlikely. However, range expansion and rapid population growth can be prevented through strategic management including pub-lic education.

Key words: Giant hogweed, Heracleum mantegazzianum, Apiaceae, HERMZ, invasive plant, weed biology, furanocoumarins

Page, N. A., Wall, R. E., Darbyshire, S. J. et Mulligan, G. A. 2006. La biologie des plantes exotiques envahissantes au Canada. 4.Heracleum mantegazzianum Sommier & Levier. Can. J. Plant Sci. 86: 569–589. Heracleum mantegazzianum (berce du Caucase) estune espèce exotique envahissante qui préoccupe les producteurs du sud du Canada, car la plante s’est échappée des jardins pour colonis-er les écosystèmes naturels, rudéraux et agricoles. En plus d’être une adventice dans les régions agricoles et urbaines, l’espèce s’avèreune menace pour les écosystèmes naturels et la santé humaine. Membre de la famille de la carotte (Apiacées), la berce du Caucase estétroitement apparentée à l’espèce indigène Heracleum maximum Bartram (berce laineuse). Cette vivace monocarpienne fleurithabituellement la troisième ou la quatrième année. Il est facile de la distinguer des autres plantes à cause de sa grande taille, de la formede ses feuilles, des taches rouge foncé sur la tige et les pétioles et des fruits très caractéristiques. On la trouve de plus en plus dans leshabitats riverains, les plaines inondables et la lisière des forêts, ou à proximité des zones urbaines dans le sud-ouest de la Colombie-Britannique et le sud de l’Ontario. D’après les spécimens conservés dans les herbiers, H. mantegazzianum a été identifiée pour la pre-mière fois en 1949 en Ontario, en 1964 en Colombie-Britannique, en 1980 en Nouvelle-Écosse, en 1990 au Québec et en 2000 auNouveau-Brunswick. Le développement de peuplements denses peut nuire à la diversité des espèces indigènes. Toucher la berce duCaucase entraîne parfois une phytophotodermatose, inflammation grave de la peau résultant de la photoactivation des furanocoumarinesprésentes dans la sève par les rayons ultraviolets. Les méthodes de lutte comprennent l’application d’herbicides, la coupe mécanique etla paissance, mais on devrait aussi adopter des stratégies pour empêcher la dispersion des semences et la germination des graines dor-mantes. La propagation de cette plante dans le sud du Canada laisse planer un doute sur la possibilité de son éradication. Néanmoins,on pourrait éviter l’expansion de son aire et ralentir la croissance de la population par des méthodes de gestion stratégiques comprenantla sensibilisation de la population.

Mots clés: Berce du Caucase, Heracleum mantegazzianum, Apiacées, HERMZ, plante envahissante, biologie des mauvaises herbes, furanocoumarines

1. Species Name and Taxonomic RelationshipsHeracleum mantegazzianum Sommier & Levier (1895) —Synonyms: H. grossheimii Manden. — giant hogweed(Darbyshire et al. 2000); cartwheel-flower (Huxley 1992);giant cow-parsnip (Morton 1975, 1978); giant Russian hog-weed (Epstein 1987); Siberian cow-parsnip (Dodd et al.1994); wild rhubarb (Camm et al. 1976a); berce du

Caucase (Darbyshire et al. 2000). Bayer code: HERMZ.Apiaceae (= Umbelliferae), carrot family (= parsley family),Apiacées (= Ombellifères).

The genus Heracleum encompasses about 70 specieswhich are found mostly in north temperate regions, particu-larly central Asia (Mandenova 1951). Of the three speciesknown in Canada, one is a widespread native species,

569

570 CANADIAN JOURNAL OF PLANT SCIENCE

Heracleum maximum Bartram (= Heracleum lanatumMichx.), and two are introduced species, H. mantegazz-ianum and H. sphondylium L. The introduced H. sphondyli-um is rare in Canada from Newfoundland to Ontario(Morton 1978; Marie-Victorin 1997; Roland and Zinck1998; Hinds 2000) and is not considered invasive. Gianthogweed, H. mantegazzianum, has been introduced in east-ern Canada and British Columbia and has become an inva-sive plant of increasing management concern throughout itsCanadian range. It is endemic to the Caucasus region of cen-tral Asia and is widespread in Europe and North Americathrough purposeful or accidental human introduction.

There are several taxonomic uncertainties in the genusHeracleum caused by the variability of distinguishing char-acters. Although treated as separate species in the followingkey, the North America native H. maximum could not bedistinguished from the highly variable Eurasian taxon H.sphondylium by Brummitt (1971). He proposed that theNorth American plants are best treated as H. sphondyliumsubsp. montanum (Gaudin) Briquet. The name H. lanatumMichx. was adopted by Cronquist et al. (1997) who werereluctant to apply the European name to our NorthAmerican plants.

2. Description and Account of Variation(a) Species Description — This description is based oncharacteristics of Canadian populations supplemented withpublished information (e.g., Tutin 1980; Tiley et al. 1996)and with measurements and observations on specimensfrom Canada. Definitions of specialized terms and detaileddescriptions of morphological characteristics in the familyApiaceae are given by Kljuykov et al. (2004). The illustra-tions (Figs. 1A–1C) are based on plants from Canadian pop-ulations. Leaf epidermal characteristics are described byArora et al. (1982) and pollen morphology by Cerceau-Larrival (1971).

Plants are monocarpic perennials or occasionally bienni-als. A large, hollow central stem develops from a rosette ofbasal leaves. Plants are typically 3–4 m tall during flower-ing, but range up to 5.5 m. The plant has a stout branchingtaproot up to 60 cm long and 15 cm in diameter at thecrown. Stems are 4–10 (–15) cm wide, hollow and ridged(Fig. 1A). Stems and lower petioles are usually extensivelyblotched or spotted with reddish-purple pigments, and withpustulate-based hollow hairs, especially near the nodes (Fig.2). Leaves are alternate. Lower leaves are 1–2.5 m long,compound, divided to a varying extent but usually in three(ternate) or sometimes five (pinnate) segments (the lowerones usually with a short stalk), which are usually deeplylobed and coarsely and irregularly toothed. The leaves areusually pubescent beneath, particularly when young, andmore or less glabrous above; petioles are stout, hollow andwith a broad sheath at the base; upper stem leaves are small-er (sometimes not divided) with longer petioles and moreinflated sheaths. Most parts of the plant, except the upperleaf surfaces, are covered with stiff whitish hairs (Fig. 2).The hairs are brittle and hollow. When fresh they are filledwith a clear sap. The plant has a strong resinous or musty

smell, particularly when the stem or leaves are brushed orbroken.

The inflorescence is a large compound umbel composedof 4–12 flat-topped or broadly rounded units, the main onesare up to 80 cm wide and with 50–150 hairy umbel rays thatare 15–40 cm long and terminate in smaller umbellets; thepeduncle is as long as or longer than the rays (up to about 60cm) and usually with spreading hairs; the terminal umbelcontains hermaphrodite flowers and is surrounded by up toeight lateral umbels, which contain mostly male flowers.Flowers are whitish; sepals triangular; petals creamy whiteor rarely pinkish, up to about 1 cm long but irregular in size,the larger ones toward the outside of umbellets and deeplynotched, the smaller ones toward the inside of umbellets areovate and not notched; styles enlarged at the base (sty-lopodium). Flowers are lightly scented.

Fruits are dry schizocarps that consist of two stronglyflattened mericarps (seeds), which are joined until ripening,6–18 mm long, 4–10 mm wide, and 1 mm thick; mericarpsare elliptical and emarginate at the apex, with slender, lowdorsal ridges and broadly winged lateral ridges, the wingsflat and closely appressed to one another, hairy on the outerface when young and becoming glabrous with age.Mericarps have 4–6 prominent abaxial oil tubes (vittae)about 3/4 as long as the fruit (Fig. 3), usually broadened(spatulate or clavate) and about 1 mm wide at the lowerend; pedicels 10–20 mm, hairy; the persistent styles aredivergent or somewhat recurved. Cotyledons are linear andnarrowed to a distinct petiole. The first leaves are simple,rounded and more or less kidney-shaped with a crenulatemargin (Fig. 4).

Chromosome counts of n = 11 and 2n = 22 were obtainedfrom plants that were grown from seeds collected atWakefield, Québec (reported here). Another gametophytecount of n = 11 was reported from plants cultivated atPortland, Oregon (Bell and Constance 1966). Wanscher(1932) counted n = 11 from material in cultivation atCopenhagen and Hindáková and Schwarzová (1987) count-ed 2n = 22 from material naturalized in Slovakia. The onlycount from plants growing in their native range in Georgiawas 2n = 22 (Gagnidze and Chkheidze 1975).

(b) Distinguishing Features — Large size, leaf shape, darkreddish pigments in patches or spots on stems and petioles,and fruit characteristics readily distinguish H. mantegazz-ianum from other species of Apiaceae in Canada. Species ofthe genus Angelica are superficially similar, but both thenative A. atropurpurea L. and the introduced A. sylvestris L.have twice pinnate principal leaves with the distinct leafletstoothed but not lobed.

Key to Heracleum species in Canada:

1. Flowering stems 3–4 m tall (up to 5.5 m), usually withextensive reddish-purple blotching; principal leavesdeeply lobed and toothed, 1–3 leaflets (1-ternate), thestalks of the lower leaflets usually less than 10 cmlong; rays of main umbel more than 50; fruits ellipticto ovate, with oil tubes broadened towards the ends

PAGE ET AL. — HERACLEUM MANTEGAZZIANUM 571

and extending 3/4 or more of the seed length (Figs. 1Band 3); introduced species in British Columbia andEastern Canada..............................H. mantegazzianum

1. Flowering stems about 2 m tall (up to 3 m), usually with diffuse reddish-purple coloured blotches or spots;principal leaves with 3–5 leaflets (ternate to pinnate),

Fig. 1. Heracleum mantegazzianum: A. Cross section through stem; B. abaxial (outside) view of the schizocarp or fruit (note the two styles)showing four dark oil tubes; C. adaxial (inside) view of mericarp or seed (half of the schizocarp).


the stalks of the lower leaflets sometimes 10 cm ormore long; rays of main umbel less than 50; fruitsovate to obovate with linear oil tubes scarcely or notbroadened at the ends and extending 1/2–3/4 of theseed length (Fig. 3)......................................................2

2. Plants with soft, more or less woolly hairs; principal leaves more or less palmately divided into threeleaflets (ternate), the lower leaflets mostly with peti-oles about 10 cm or more long; fruits deeplynotched; native species widespread throughoutCanada ..................................................H. maximum

2. Plants with stiff hairs; principal leaves with 3–7 leaflets (once pinnate), the lower leaflets sessile

or with short stalks; fruits shallowly notched;uncommon and local introduced species in EasternCanada ............................................H. sphondylium

(c) Intraspecific Variation — Little intra-specific variationhas been detected in H. mantegazzianum and no subspecifictaxa have been described. In a chromatographic study of 18plants from a Scottish population, Weimarck et al. (1979)found only two phenotypes, as opposed to eight phenotypesin the native H. sphondylium, of which one was found in94% of plants. Walker et al. (2003) used microsatellite DNAmarkers to assess genetic diversity in 13 H. mantegazzianumpopulations in three drainage catchments in Britain. Theyfound that overall genetic variation was high, and that pop-

Fig. 2. Sheathing base of petiole surrounding the flowering stem. Note the long white hairs with pustulate reddish bases.


ulations from the same catchment were more similar, sug-gesting connections through waterborne dispersal.

3. Economic Importance and EnvironmentalImpact(a) Detrimental — Considered to be one of the most noxiousinvading plants in Europe (Pysek et al. 1998), H. mantegazz-ianum poses a threat to natural ecosystems and human health,as well as being a weed in agricultural and urbanized areas. InEurope it has rapidly established in a variety of semi-naturaland man-made ecosystems, particularly floodplains, riparianzones, forest edges, roadsides, meadows, open forest, andunmanaged urban areas (Williamson and Forbes 1982; Tileyand Philp 1992; Pysek 1994; Pysekand Prach 1994; Otte andFranke 1998). It often forms monospecific stands (Fig. 5)where its tall stems and large leaves effectively compete forlight against other plants (Clegg and Grace 1974; Williamsonand Forbes 1982; Pysek 1991; Andersen 1994; Otte andFranke 1998). There is some evidence of allelopathy inHeracleum species (Junttila 1975, 1976), which may increasethe detrimental impact of H. mantegazzianum on other plants.In general, H. mantegazzianum is believed to reduce diversityof native plant communities (Tiley and Philp 1992; Godefroid1998), although there is little comprehensive research to assessthis impact. It is especially invasive in riparian ecosystems,where new colonies can be established from waterborne seeds(Dawe and White 1979; Pysek 1994). It can displace riparianvegetation and increase streambank erosion during the winter

when H. mantegazzianum is senescent (Wright 1984; Tileyand Philp 1992, 1994; Dodd et al. 1994). Instability of riverbanks dominated by H. mantegazzianum in Great Britain andIreland poses a serious threat to salmon spawning habitats(Caffrey 1999).

It is well known that H. mantegazzianum causes phy-tophotodermatitis, a serious skin inflammation resultingfrom the activation, under ultraviolet radiation, of com-pounds contained in the plant sap (Kuske 1940; Drever andHunter 1970; Camm et al. 1976a). The major phytotoxicprinciples in Heracleum species are linear furanocoumarinsor psoralens, mainly 5-methoxypsoralen and 8-methoxypso-ralen (Pathak et al. 1967; Nielsen 1970; Molho et al. 1971;Pira et al. 1989). Toxicity is due to cross-linking of the furanring with pyrimidine bases (thymine) of DNA in the pres-ence of ultraviolet light at 315–400 nm. Furanocoumarin-induced dermatitis typically consists of painful blisters thatform within 48 h and become pigmented scars that can lastas long as 6 yr, but more typically disappear after severalmonths (Sommer and Jillson 1967; Morton 1978; Tiley et al.1996). Long-term sensitivity of affected skin areas to sun-light may follow. Furanocoumarins may also be mutagenic(Igali et al. 1970; Townsend et al. 1971; Clarke 1974).Numerous cases of phytophotodermatitis attributable tocontact with H. mantegazzianum have been documented bymedical practitioners (Miescher and Burckhardt 1937; Jonesand Russell 1968; Smellie 1968; Drever and Hunter 1970;Camm et al. 1976a; Mitchell and Rock 1978) and many

Fig. 3. Abaxial (outside) surface of Heracleum mericarps showing differences in oil tubes (vittae): A. H. maximum; B. H. mantegazzianum.


more have been reported by the public media (e.g.,Kamermans 1977; Anonymous 2002; Leslie 2003; Dutton2004). In addition, injuries to livestock and pets have beenattributed to furanocoumarins in H. mantegazzianum(Hinterman 1962; Camm et al. 1976a; Andrews et al. 1985;Harwood 1985). However, Dickson (2001) noted thatdespite the widespread presence of H. mantegazzianum inGlasgow, Scotland, local dermatologists considered phy-tophotodermatitis to be an uncommon medical problem.

Although H. mantegazzianum is not commonly regardedas an agricultural weed, there is some concern about itsestablishment in pastures and field crops because of injuriesto livestock. In Sweden it has been found to invade arablelands such as potato fields (Lundström 1984). Some of theseconcerns have been offset by reports of successful suppres-sion of H. mantegazzianum through foraging by livestock(Andersen 1994; Tiley et al. 1996). It can also harbour pestsand pathogens of agricultural crops (see Section 13).

There is a possibility of detrimental effects to H. maxi-mum in Canada from hybridization with H. mantegazz-ianum. Occurring in a variety of habitats across NorthAmerica, H. maximum will be increasingly in contact withH. mantegazzianum as the latter spreads. Although H. max-imum flowering peaks 3–4 wk earlier than H. mantegazz-ianum, there is sufficient overlap in flowering times forcross-pollination to occur. In Europe, H. mantegazzianum isknown to hybridize with H. sphondylium (see Section 9).

(b) Beneficial — The striking size and distinctive inflores-cences make H. mantegazzianum an impressive ornamentalplant for gardens (Mondoor 2001). Indeed, some of its rapidspread in Canada is likely caused by purposeful introductionsfor horticulture. As of 2005, it is still seen for sale in someCanadian nurseries (Darbyshire, personal observation). Itspopularity as an ornamental plant has been identified as amajor cause of its spread in central Europe (Pysek 1991).

The flowers are nectar sources for honeybees (Grace andNelson 1981; Bürki and Nentwig 1997) and the driedumbels are used for decorative purposes. In addition, theseeds are used as a spice (golpar) in Middle Eastern cooking(Westbrooks 1991), although five Heracleum species nativeto mountainous areas of Iran are traditionally the source ofgolpar. Both H. mantegazzianum and its close relative H.sosnowskyi Manden. have also been cultivated as forageplants in Eastern Europe (Nielsen et al. 2005). Nutrient con-centrations were examined by Otte and Franke (1998) andcompared favourably with other forages. Furanocoumarinsgathered from plant sources (including Heracleum species)are used extensively in the treatment of leukodermia and inthe preparation of sun-tan lotions (Devagiri et al. 1997).They have also been found to have potential as insect repel-lents as they suppress feeding and growth in some species(Klocke et al. 1989). Various compounds present in H. man-tegazzianum, including furanocoumarins and glycosides,

Fig. 4. Four seedlings of Heracleum mantegazzianum about 1 wk after emergence.


have been investigated for their anti-bacterial and anti-fun-gal properties (Fischer et al. 1976, 1982).

(c) Legislation — There is no federal or provincial legisla-tion preventing entry of H. mantegazzianum into Canada ordesignating it as a noxious weed (Darbyshire 2003). TheDistrict of Saanich on southern Vancouver Island, BC,recently designated it a noxious weed under a municipalbylaw (District of Saanich 2005). In Ontario, it is listed as anoxious weed in the bylaws of Grey (3379-92) and Huron(24-92) counties, where it is to be controlled on agriculturallands under section 10 of the Ontario Weed Control Act (M.Cowbrough, personal communication).

In the United States, H. mantegazzianum is listed underthe Federal Noxious Weeds Act of 1974 whereby it must bereported and controlled when found and importation orinterstate movement is illegal (Anonymous 1999).Washington State lists it as a Class A weed (Anonymous2003a) and prohibits sale and transport of propagating mate-rials (Hamel and Parsons 2001). In the United Kingdom,there is a ban on planting H. mantegazzianum under the1981 Wildlife and Countryside Act (Anonymous 2005)

4. Geographical DistributionThe natural range of H. mantegazzianum is the subalpinezone of the western Caucasus Mountains of Georgia,

Azerbaijan, and southern Russia (Mandenova 1951; Otteand Franke 1998), where it is found in meadows, clearings,and forest edges between 1500 and 1850 m.

Since the 19th century, H. mantegazzianum and a number ofother Caucasus plants have been introduced to a variety ofregions of both the northern and southern hemispheres, partic-ularly Europe (Clegg and Grace 1974; Knapp and Hacker1984; Pysek 1991). Early range expansion of H. mantegazz-ianum appears to have begun in the early to mid-19th centurywhen it was introduced to botanical gardens in the UnitedKingdom and continental Europe as an ornamental plant.Introduction to North America likely occurred during the firsthalf of the 20th century.

In Canada, it is found from British Columbia toNewfoundland (Fig. 6A). In British Columbia, it is mostprevalent on southeastern Vancouver Island, particularly theParksville area, and in North and West Vancouver (Fig. 6B)(Page and Wall 2003). It is also widespread in southwesternOntario, with most occurrences found east of Lake Huron(Fig. 6A). Localized populations are known in southernQuebec, New Brunswick, Cape Breton and easternNewfoundland. It has been recorded in six US states, eitherin the northeast or Pacific northwest: Conneticut, Maine,Massachusetts, Michigan, New York, Pennsylvania,Oregon, and Washington (Kartesz 1999)

Fig. 5. A large colony of Heracleum mantegazzianum growing in a roadside area in North Vancouver, BC.


It has also been reported as established in Austria,Belgium, Czech Republic, Denmark, Finland, France,Germany, Hungary, Iceland, Ireland, Italy, Latvia,Luxembough, the Netherlands, Norway, Poland, Slovakia,

Sweden, Switzerland, and the United Kingdom (Brummitt1968; Hindáková and Schwarzová 1987; Nielsen et al.2005) and in New Zealand, mainly eastern South Island(Webb et al. 1988).

Fig. 6. Distribution of Heracleum mantegazzianum as compiled from herbarium specimens (circles) and sight records (triangles). Herbariumspecimens examined from CAN, DAO, HAM, MICH, MTMG, NSPM, OAC, QUE, TRT, UBC, UNB, UWO and V (acronyms afterHolmgren et al. 1990) A. Canadian distribution; B. Detail of distribution in southwestern British Columbia.


5. Habitat(a) Climatic Requirements — Although it grows in a widerange of climatic conditions, H. mantegazzianum is mostinvasive in regions with cool, moist climates that are similarto its native habitat. In Canada, its distribution is primarilyin developed areas and the limiting effects of climate on itsrange are unknown. It does, however, occur as far north asKapuskasing, Ontario (Fig. 6A). In the Czech Republic,Pysek et al. (1998) found that peak incidence of H. man-tegazzianum was associated with January isotherms of 3°C,June isotherms of 7°C, and annual precipitation of 400–500mm. January isotherm and density of human populationwere the factors most significantly correlated with H. man-tegazzianum distribution. In the Scandinavian countries, H.mantegazzianum is found only in Denmark, southernSweden and southern Norway (Nielsen et al. 2005). Basedon the climates in which this species grows, it is readilyapparent that overwintering plant parts (e.g., roots and stem-bud) can withstand cold conditions. Plants, includingseedlings, are frost resistant (Godefroid 1998) and sproutingshoots have been observed to survive –17°C (Tiley et al.1996). Its occurrence in northern Ontario (Fig. 6A) and itsdistribution in northern European countries (Nielsen et al.2005) suggest it could be tolerant of the extremely cold,continental climates in Canada.

(b) Substratum — The prevalence of H. mantegazzianum inriparian zones, ditches, and forest edges suggest that itprefers moist, fertile soils, but this may be due largely to itstendency to spread along streams (Pysek and Pysek 1995;Tiley et al. 1996; Otte and Franke 1998). Large colonies areoften found on sandy or silty deposits along watercourseswhere seeds come to rest (Morton 1978; McIlveen 2001),but seedlings may also become established in organic soilstypical of woodlands. Deep soils are required for optimumgrowth, but plants may also thrive on drier, well-drainedsoils characteristic of roadsides and waste-ground(Godefroid 1998). Along French Creek on VancouverIsland, BC, initial establishment has been in alluvial soilsalong the stream, but there has been subsequent spread intoadjacent woodlands and along road embankments (Wall,personal observation). This distribution pattern was alsoobserved in North Vancouver, BC (Page, personal observa-tion). Morton (1978) reported it growing in deep rich moistsoils in semi-shade, mostly along rivers and streams, inOntario. In eastern Ontario and western Québec, plants arefrequently growing in sandy-gravel of road embankmentsand ditches (Darbyshire, personal observation).

In the United Kingdom, H. mantegazzianum has beenfound on a wide variety of soils ranging from clays to grav-els, and from limestone to acid heath, but high organic mat-ter (usually 3–9%) and high potassium and calcium levelsare especially favourable (Clegg and Grace 1974). Tiley andPhilp (1992) report occurrences on sandy soils in riparianand ruderal habitats. Soils supporting H. mantegazzianumare usually neutral to alkaline; a pH range from 5 to 9 hasbeen reported in Scotland (Clegg and Grace 1974). Pysekand Pysek (1995) reported that it is not found in acidic habi-tats in the Czech Republic. Detailed soil analysis at a site in

Germany by Otte and Franke (1998) also showed highorganic matter (1.5–4.9% organic carbon) and slightlyacidic pH (6.4–6.7). The presence of H. mantegazzianum onsea shores and oceanside cliffs with frequent salt spray, aswell as roadside areas receiving de-icing salt, indicates thatit is salt tolerant (Tiley et al. 1996; Godefroid 1998).Lundström (1984) also reported that it occurs in coastalsands in southern Sweden.

(c) Communities in which the Species Occurs — In Canada,H. mantegazzianum occurs in open and forested floodplains,alluvial gravel bars, roadside ditches and embankments,hedgerows, open forests, forest edges, marine shorelines,and old fields (Morton 1978; McIlveen 2001; herbariumspecimen label data). The pattern of H. mantegazzianumdistribution in Canada suggests that initial establishment isusually in disturbed, nonforested habitats such as roadsides(see Fig. 5), forest edges, and stream margins and from thesefoci it can invade open woodlands, grasslands or wet mead-ows. H. mantegazzianum does not thrive and spread indense coniferous forests, grasslands, or sedge meadowswhere competition from the existing community may pre-clude seedling establishment or growth. It is also intolerantof wetlands subject to long periods of inundation. On allu-vial or fluvial sites, H. mantegazzianum often grows inmixed communities with herbs and grasses, but it canbecome dominant on drier, disturbed sites.

In its native range, H. mantegazzianum occurs in subalpinemeadows, ravines, along forest edges, and along streams(Mandenova 1951; Tiley et al. 1996; Otte and Franke 1998;Wittenberg et al. 2003). In regions where it has recently beenintroduced, it occurs mainly in riparian areas and near humanhabitations where it has been planted or has escaped from gar-den cultivation. Populations in Europe are frequently distrib-uted along dispersal corridors, such as watercourses, roadsidesand railways (Pysek 1991; Otte and Franke 1998), althoughmany types of communities have been colonized (Pysek andPysek 1995) and its ecological range is still expanding(Caffrey 1999). A similar pattern was found in a surveyaround Edinburgh, Scotland (Clegg and Grace 1974), and inBelgium it grows principally along roads and railways and inwaste ground and parks (Godefroid 1998). In regions where H.mantegazzianum has been present for many decades, it occursin a broader range of habitats including riparian areas, road-sides, pastures, open woodlands, wetlands, landfills, urbanparks, and urban gardens (Andersen 1994; Pysek 1994; Tileyet al. 1996). Otte and Franke (1998) state that it is a plant ofnitrophitic herbaceous perennial communities in centralEurope where it occurs in a variety of sites including aban-doned farmland, eutrophied forest edges, and transportationcorridors in lowland and submontane regions. They analysedthe plant communities at a number of sites in the Lahn Valleyin central Germany and found that H. mantegazzianum formeda distinct community with Galium aparine L. and Urticadioica L.

6. HistoryThe introduction of H. mantegazzianum into Canada wasprobably as a result of escape from horticultural planting


(Morton 1978), as it was in many parts of Europe. The ear-liest herbarium specimen of H. mantegazzianum in Canadawas collected from Kemble, ON in 1949. The first widelydistributed literature report of its occurrence in Canada wasthat of Morton (1975) which was based on his observationson a population at Tara, ON. Additional surveys by Morton(1978), after his initial discovery, indicated that it was wide-ly spread in eight counties in southwestern Ontario. The ear-liest specimen from British Columbia was collected in 1964near Parksville on southeastern Vancouver Island, althoughit may have been introduced to the province in the 1920s or1930s (Merilees 1978; Dawe and White 1979; Merilees1981). The first specimen from the Lower Fraser Valley insouthwestern British Columbia was collected in 1974 fromMahon Park in the City of North Vancouver, BC. In otherprovinces the first collections were made at Baddeck, NS, in1980; Stanstead, QC, in 1990; and, Sackville, NB, in 2000.

7. Growth and Development(a) Morphology — Various morphological features in com-bination with its toxic properties make H. mantegazzianumvery difficult to control. The sheer size of older rosettes andmature plants, as well as the usually dense growth, posechallenges for control practices. Large taproots store con-siderable below-ground resources that confer some protec-tion against physical and chemical control measures. Aswell as providing energy reserves in the event of poor grow-ing conditions, the large taproots are difficult to dig up andare resistant to energy depletion through repeated cutting ofabove-ground parts.

The large root reserves also allow plants in the vegetativegrowth stage to develop an extensive leaf canopy elevatedon long petioles early in the growing season. This facilitatescompetition with other herbaceous species and results inhigh productivity. In Germany, Otte and Franke (1998) cal-culated the total biomass production at 947 t ha–1 and theabove-ground biomass at 710 t ha–1.

All parts of the plant contain toxic furanocoumarins (seeSection 3a), which is a hazard for those attempting vegeta-tion management. The lower parts of the hollow stems andpetioles may be partly filled with fluid containing fura-nocoumarins, sometimes in considerable volumes(Darbyshire, personal observation). Mechanically cuttingthe plant risks splashing or volatilizing sap and fluids,putting workers at high risk. The hollow hairs, often abun-dant on the stems around the nodes and on the lower petioles(Fig. 2), also contain sap. These brittle hairs may penetrateand break off in skin causing physical irritation (Morton1978), as well as facilitating delivery and penetration offuranocoumarins.

(b) Perennation — H. mantegazzianum is a monocarpicperennial or occasional biennial. In Germany, plants mayflower in the second year, but usually mature after 3 or moreyears (Otte and Franke 1998). Above-ground plant partsannually die back to the taproot each autumn until the yearthat flowering occurs, after which the whole plant senesces.Plants may flower in the 2nd, 3rd, 4th or 5th years but mostflower in the 4th year in Ontario (Morton 1978) and the 3rd

year in the United Kingdom (Stewart and Grace 1984; Tileyet al. 1996).

There may occasionally be continued growth after flow-ering, and Morton (1978) reported that some plants producenew crowns after flowering. Similarly, a second floweringstem is usually produced if the main stem is cut or damagedbefore the flowers open (Stewart and Grace 1984; Dodd etal. 1994).

(c) Physiological Data — Since the phloem of H. man-tegazzianum is easily separated from the vascular bundles, ithas been used extensively in studies of translocation.Fensom et al. (1968) estimated a streaming rate of about 1cm h–1. Other studies have been conducted, mainly at MountAlliston University in Sackville NB, using H. mantegazz-ianum as a research tool to elucidate the underlying forcesgoverning mass flow (Tyree and Fensom 1970; Spanner1970), determine the nature of particles observed in phloem(Lee et al. 1971; Hart and Sabnis 1975; Barclay et al. 1977;Murphy 1986), and gain further insight into energy transfersin active phloem (Ezeala et al. 1974).

The chemistry of H. mantegazzianum is well studied becauseof its production of complex secondary compounds, particu-larly furanocoumarins (Murray 2002), but also flavonoids(Harborne 1971), glycosides (Fischer et al. 1982), and essen-tial oils (Jain 1969). A considerable amount of information issummarized in the comparative phytochemistry study byCarbonnier et al. (1982). Furanocoumarin chemistry and con-tent vary between plant parts and growth stages. BothKnudsen (1983) using material from Denmark and Pira et al.(1989) using material from Italy found maximum levels in thefruit, with intermediate levels in the leaves, followed by theroots, and the lowest levels in the stem. Psoralen was the pre-dominant furanocoumarin in the leaves and bergapten in thefruit (Pira et al. 1989). Furanocoumarin levels peaked in springand early summer and declined in late summer and autumn.Glowniak et al. (2003) isolated bergapten, xanthotoxin,isopimpinellin, imperitorin, pimpinellin from the fruits of H.mantegazzianum collected in Poland. Analysis by Dragan etal. (1999) of material collected in France found the same com-pounds, as well as byakangelicol.

(d) Phenology — First year growth is usually slow, and theseedling initially develops a single leaf or a rosette of foursmall, sessile leaves in early spring (Fig. 4). These are fol-lowed by larger, more petiolate leaves. A narrow taproot,45–60 cm deep, develops quickly and the upper part of theroot thickens as food reserves accumulate. In the juvenilevegetative growth stages, photosynthetic production is pref-erentially directed to the taproot for storage, and the root toshoot dry mass ratio is about 1.4:1 (Otte and Franke 1998).The root over-winters with large terminal buds that form thenext year’s shoots.

In the second and subsequent years, growth may begin asearly as late December in regions with mild winters such ascoastal British Columbia. Vegetative growth begins as agreening and swelling of buds on the root crown. By Aprilor May, plants may be 1–2 m high with three to four leaves.


Lower leaves senesce, resulting in a fairly constant leafnumber until die-back in September or October.Occasionally, 1–3 basal leaves develop in late September orOctober and remain green into the winter. Spring leaves arebroader with more surface area, while summer leaves arelonger, with narrower segments (Tiley et al. 1996; Willisand Hulme 2002).

Plants destined to flower in the ensuing year have a cer-tain threshold root size, have woody secondary thickeningof the roots, are the earliest to begin growth, and have thelargest spring leaf development. In an even-aged stand, onlythe largest plants flower and these plants usually possessthree to four large basal leaves plus four to six stem leaves.In early June, flowering plants form a swollen terminal bud,which encloses the terminal umbel. Secondary umbels formlater and may not develop fully unless the terminal umbel isdamaged (Tiley et al. 1996). Flowering occurs over a 5–6wk period between June and August (Mandenova 1951;Morton 1978; Otte and Franke 1998). The fruits mature anddry over August and September, and split into two meri-carps when ripe. These usually detach over autumn and win-ter but are sometimes retained until the following year.

Detailed studies on the growth and phenology of H. man-tegazzianum in Germany were undertaken by Otte andFranke (1998) and in Ireland by Caffrey (1999). Otte andFranke (1998) described four development phases in plantsflowering that year in central Germany: (i) vegetative devel-opment (10 wk: Mar. 01–May 10); (ii) flowering (8 wk:May 10–Jul. 03); (iii) fruiting (5 wk: Jul. 03–Aug. 10); and,(iv) release of mericarps (>4 wk: after Aug. 10). Caffrey(1999) found similar growth and flowering patterns in Irishpopulations: plants emerged in early February, grew vegeta-tively until early May when flowering began, floweringpeaked in late June, and most seeds had ripened and dis-persed by the end of August.

(e) Mycorrhiza — No records of mycorrhiza on H. man-tegezzianum have been reported. In Germany, Kühn et al.(1991) surveyed a field abandoned from maize productionfor the presence of mycorrhizal relationships. Although H.mantegazzianum was not present, they reported that H.sphondylium was occasionally infected with mycorrhizalfungi.

8. Reproduction(a) Floral Biology — H. mantegazzianum produces a largenumbers of hermaphrodite flowers on upright umbels.Flowers mature in a centripetal sequence and the primaryterminal umbel matures first (Stewart and Grace 1984; Tileyet al. 1996). Secondary or tertiary umbels, some of whichare staminate, may only mature if the terminal umbel isdestroyed (Tiley et al. 1996). Stamens mature and dehiscebefore the pistils are receptive, although self-pollination ispossible between flowers in umbels at different stages ofmaturity on the same plant (Weimarck et al. 1979; Stewartand Grace 1984). In the staminate phase, flowers of anumbel open over a 5-d period, the stamens lasting for about

a day, after which a period of 1–3 d usually elapses beforeall the stigmas across the umbel become receptive on thesame day (Stewart and Grace 1984).

There is little information on insect pollinators of H. man-tegazzianum; however, like most Apiaceae, it attracts a widerange of unspecialized pollinators. A long list of coleopter-an, dipteran and hymenopteran visitors is given for theEuropean H. sphondylium by Knuth (1908). Grace andNelson (1981) found that H. sphondylium and H. man-tegazzianum had attracted distinct insect faunas in areas ofsympatry in Scotland. They list 48 species of insects caughton H. mantegazzianum, but found significant amounts ofpollen on only 16 taxa (10 diptera, five hymenoptera andone coleopteran). Otte and Franke (1998) report that polli-nation in central Europe is primarily by beetles(Coleoptera).

(b) Seed Production and Dispersal — The large inflores-cence of H. mantegazzianum produces prodigious amountsof seed. In central Europe, each of the several large umbelsmay produce 5000 to 6500 seeds (Pysek 1991; Otte andFranke 1998). In the Czech Republic, Pysek and Pysek(1995) reported that up to 107 000 seeds may be producedby a single plant. Tiley and Philp (1994) estimated 60 000flowers (producing up to 120 000 seeds) on one large plantexamined in Scotland. Otte and Franke (1998), however,observed that not all flowers in secondary and tertiaryumbels on plants growing in Germany produced viableseeds and that seeds in these umbels are lighter and smallerthan those of the primary umbels.

Plants from Scotland studied by Weimarck et al. (1979)showed a seed set of 84% and pollen stainability (cottonblue) of 91%. In their artificial pollination experiments,Stewart and Grace (1984) found that mean seed set wasminimal when flowers within a single umbel were used(1%), but 68% when flowers between different umbels onthe same plant were used. Mean seed set in flowers betweendifferent cross-pollinated plants was 64% (range 47–85%).Germination after 22 wk at 2°C stratification was less forseed from self-pollinated than cross-pollinated flowers, 27%versus 59%, respectively.

The fruit splits into two flat and winged mericarps eachup to 20 mg in weight and possessing distinctive oil tubes(Figs. 1B, 1C, and 3). The light, buoyant fruit may be water-borne for long distances and also carried short distances bystrong air currents (Tiley et al. 1996). Dispersal occurs bydirect seed fall within the immediate vicinity of matureplants, by wind, by water along streams, ditches, or roadmargins, and by direct human transport (Otte and Franke1998; Dawson and Holland 1999). Otte and Franke (1998)found that the 75% of seeds fell within 120 cm of the moth-er plant. Several authors have mentioned that seeds may bedispersed by wind (e.g., Lundström 1990; Godefroid 1998).Although the seeds are flat with winged margins, experi-ments on wind dispersal suggest that this is not a very effec-tive means of dispersal since seeds released from a height of0.69 m, at wind speeds of 0.6–5.5 m s–1, traveled less than1 m from the release point (Clegg and Grace 1974). Otte andFranke (1998) noted that the maximum dispersal distance by


wind was 8–10 m. Wind currents from fast moving vehiclesmay contribute to the dispersal of seed along highways andrailroads (Clegg and Grace 1974). The first reported popu-lation in Michigan (Case and Beaman 1992) occurred in avery isolated locality that was clearly not a deliberate plant-ing. Its long distance dispersal to the reported location couldnot be explained by the discoverers.

Since seeds can float for up to 3 d (Morton 1978), wateris a major agent of local and long distance seed dispersal(Pysek and Prach 1994; Walker et al. 2003). Floatationexperiments in Scotland showed that seeds would remainafloat for 1.5–2 d in turbulent water and 3 d in calm water(Clegg and Grace 1974). At a modest flow rate of 0.1 m s–1,seeds would be able to travel 10 km or more. Movement ofseed-contaminated soil by humans is an important dispersalmechanism in North Vancouver, BC (Page, personal obser-vation). Movement of young plants or seeds for horticultur-al plantings is also a source of new populations. Noevidence has been found of significant transport by birds,mammals, or insects (Otte and Franke 1998).

(c) Seed Banks, Seed Viability and Germination — A seedbank study at seven sites in the Czech Republic by Krinke etal. (2005) found that the numbers of seeds in the seed bankwas directly related to plant density and that the numbersand proportion of viable seeds at any one site variedbetween years. Pooled results showed a mean total soil seedpopulation per m2 of 6719 ± 4119 in autumn, 4907 ± 2278in spring and 1301 ± 1036 in summer, 95% of which werepresent in the upper 5 cm. However, the population of viableseeds was 3759 ± 2906, 2044 ± 1198 and 192 ± 165, respec-tively.

Seeds may persist in the soil for 5–6 yr before germinat-ing (Andersen 1994) and can remain viable in the soil for upto 15 yr (Andersen and Calov 1996). However, Otte andFranke (1998) found that seed-banks were minimal for pop-ulations in Germany. Morton (1978) found that seeds fromOntario plants stored at room temperature were still viableafter 7 yr, although Lundström (1990) stated that seeds mayremain viable for as much as 15 yr. Germination success ofolder seeds was tested on seed obtained from herbariumspecimens by Hallam and Dodd (1996). Seeds ranging from2 to 67 yr old were placed on moistened filter paper for 16wk and then transferred outdoors for 1 mo. Two-year oldseeds germinated at a rate of 78%, but all older seeds failedto germinate.

In British Columbia, germination usually occurs duringlate winter and early spring (Wall, personal observation). InOntario and Québec, seeds are often seen on dead stemsfrom the previous year. In Quebec, seeds collected fromdead stems of the previous year showed 50–80% germina-tion at room temperature in laboratory conditions(Darbyshire, unpublished data). In coastal British Columbia,seeds collected the previous autumn, stored over winter inan unheated shed and planted in late March or early April inpotting soil, germinated between April and June, ceased ger-mination during late June and July, and resumed the follow-ing spring after dry storage (Wall, unpublished data). Aperiod of cold temperature of 2–3 mo is required for optimal

physiological development and embryo maturation in H.sphondylium (Stokes 1952), and although definitive experi-ments have not been done with H. mantegazzianum, similarprocesses are likely present in that species. Results of ger-mination tests in H. mantegazzianum have varied consider-ably but general conclusions are that seeds are dormant orrequire a long after-ripening period after being shed. Krinkeet al. (2005) identified dormant seeds in Czech populationsby testing the viability of non-germinating seeds with tetra-zolium. Seeds that were artificially dried and stored indoorsoften failed to germinate, but seed stored outdoors through-out the winter germinated well (Tiley et al. 1996). Coldstratification at 0–5°C for 6–12 wk improved germination(Grime et al. 1981; Baskin and Baskin 1998). In germina-tion tests of 28 d duration by Andersen and Calov (1996),germination averaged 25% in seeds that were subjected to atemperature of –18°C for 3 wk, which was higher than seedsstored at room temperature. In central Germany, germina-tion occurred in early spring only after seeds were exposedto 2–3 mo of cold temperatures (Otte and Franke 1998).Although no light requirements have been demonstrated forgermination (Moravcová et al. 2005), best emergence in thefield has been observed in the open or under light vegetativecover in the United Kingdom (Tiley et al. 1996). Most seedsgeminate from a depth of less than 5 cm, although emer-gence from 10 cm has been observed (Krinke et al. 2005;Moravcová et al. 2005). Seeds will rot in flooded soil.

In the Czech Republic, laboratory studies by Moravcováet al. (2005) found the mean germination rate of greater than90% under optimal conditions. Seedling mortality is, how-ever, also high. In Ireland, germination begins in Februaryand peaks in April. Only 1.2% and 13.7% of seedlings pre-sent in April were extant by the end of August in the twopopulations studied by Caffrey (1999). Tiley et al. (1996)estimated that less than 23% of germinated seedlings sur-vived to maturity. No study of mortality factors wasreviewed but since most of the 20–100 000 seeds producedper plant fall within a few metres of the parent plant (Otteand Franke 1998), seedlings develop under very crowdedconditions and mortality is expected to be high.

(d) Vegetative Reproduction — Although reproduction isusually considered to be entirely by seed (Tiley et al. 1996;Otte and Franke 1998), vegetative reproduction from peren-nating buds on tuberous rootstocks has been reported(Anonymous 2003a).

9. HybridsHybridization is unusual between species in the familyApiaceae, but occasional hybrids between H. mantegazz-ianum and H. sphondylium have been reported in Europewhere both species grow in proximity (Weimarck et al.1979; Tutin 1980; Grace and Stewart 1982; Stewart andGrace 1984; Ochsmann 1996). Fertility of hybrids is low(seed set less than or equal to 1%; pollen stainability lessthan or equal to 7%) and introgression has not been detect-ed (Weimarck et al. 1979; Stewart and Grace 1984),although one plant analysed by Weimarck et al. (1979) mayhave been a backcross. Stewart and Grace (1984) found that


the prevalence of hybrids between H. mantegazzianum andH. sphondylium in the United Kingdom was reduced by thelack of habitats shared by the species, and by low viability(<10%) of hybrid seed. They also found that artificial F1hybrids were morphologically intermediate between the par-ents, had abnormal flower and fruit development andshowed structural weakness in stems. Hybrids had a greatertendency to remain vegetative, and like H. sphondylium,were not monocarpic. Epidermal characteristics of theleaves, including hairiness and stomatal patterns, were inter-mediate between the parental species on those plants con-sidered to be hybrids (Arora et al. 1982). Experimentalcrosses of H. mantegazzianum and H. sphondylium inScotland have been successful only when the pistillate par-ent was H. sphondylium (Steward and Grace 1984).

The presence of H. sphondylium × H. mantegazzianumhybrids in the field, the ease of artificial hybridization, andthe absence of specialized pollination mechanisms suggeststhat H. mantegazzianum may hybridize with the NorthAmerican H. maximum, which is part of the circumboreal H.sphondylium species complex. However, no hybrids havebeen reported in Canada.

10. Population DynamicsThe establishment and rapid spread of H. mantegazzianumdepends on dispersal of seeds both locally and regionallyfrom founder populations. Founder populations are general-ly very small, often a single plant in garden cultivation, andinitial population increases are slowed by the requirementfor 2–5 yr of growth before flowering and seed productionoccur. Early spread from founder populations may also beimpeded by the lack of nearby routes, such as watercourses,for rapid dispersal into the surrounding landscape. As Otteand Franke (1998) documented, seed dispersal in theabsence of water, wind, or soil movement is poor. However,once the population has expanded into riparian areas,regional population increase can occur rapidly throughwaterborne dispersal (Pysek 1994; Wadsworth et al. 2000;Page, personal observation).

The spatial pattern of H. mantegazzianum in NorthVancouver, BC, provides a useful model to characterizepopulation dynamics during initial invasion. Introduction ofH. mantegazzianum to this area occurred approximately 80yr ago (Merilees 1981). A recent assessment found that thepresent population covered a 0.7 ha area distributed across104 sites; only one population was larger than 250 m2 in totalpatch size and contained approximately 450 plants (KerrWood Leidal Associates Ltd. 2005). The largest populations(>100 plants) were found in riparian areas and forest edgesnear what is believed to be the epicenter of introduction withthe smallest populations occurring at the margin of rangeexpansion (Fig. 5). Most populations were strongly connect-ed by dispersal pathways, particularly water movement, andmany small populations could be traced to the movement ofseeds by water, wind, or soil from nearby populations. Veryfew populations were found in habitats away from dispersalroutes which suggests that population development is still inthe lag phase (Wade 1997). Outside of North Vancouver,populations in the Lower Fraser Valley are small and repre-

sent initial founder populations in parks and residential gar-dens. They range from 1 to 20 plants, although most consistof two to three plants in a small group. Isolated plants in res-idential gardens are increasingly common in the region (Page,personal observation).

Populations of H. mantegazzianum were studied in asmall drainage system in Halton County, ON, by McIlveen(2001). Although population size at various sample pointswas affected by many variables, he found a trend in reducedpopulation size in a downstream direction. In one streamsystem the change was from >2000, 2, 170, 138, 350, 11, 2,0, 0, 0, 4 and 0 plants and in another the change was 79, 19,32 and 0 plants. Based on the population presence withinfloodplains and the smaller downstream populations,McIlveen concluded that H. mantegazzianum was dispersedalong these watercourses by water. Based on different pop-ulation sizes and the timing of population discoveries, hespeculated that the construction of water control structureshas probably reduced the rate of spread.

In central Germany, Otte and Franke (1998) found that H.mantegazzianum populations were characterized by highdensities of seedling germination in open areas wheremature plants had flowered and died the previous year.There was competition within seedling cohorts for resourcesand high mortality in the first few months of growth. Plantsin the vegetative growth phase that were more than 1 yr oldhad a maximum of 5–7 leaves in a rosette at the root crown.These plants formed a canopy 1.6 m high by mid-June.Similar observations were made on populations in Irelandby Caffrey (1999).

Pysek (1994) and Pysek and Pysek (1995) reconstructedthe spread of H. mantegazzianum in the Czech Republic.They documented an 80-yr lag phase between initial estab-lishment and the onset of an exponential growth phase.During the lag phase, its spread was associated with its cul-tivation as a garden ornamental and subsequently alongwatercourses and other dispersal routes (Pysek 1994).During the exponential phase (post-1945), populationsdeveloped away from dispersal-related habitats and popula-tions became common throughout the general landscape.Population growth in the British Isles has followed a similarpattern (Tiley et al. 1996; Collingham et al. 2000).Wadsworth et al. (2000) modelled the rate of spread of H.mantegazzianum in a single watershed and found there wasa lag phase of 10–25 yr before exponential growth in occu-pancy of suitable habitats occurred.

11. Response to Herbicides and Other ChemicalsNo reports of herbicide tests have been found for Canada,but work in other countries indicates that H. mantegazz-ianum is sensitive to most commercial herbicides includingglyphosate, imazapic, trichlopyr, dicamba, 2,4-D, andclopyralid. Treatments usually need to be repeated annuallyand, in some situations, within the same growing season.

In Washington State, 100% top kill was achieved with anearly spring application of glyphosate at 1.7 kg a.i. ha–1 insingle plant tests, followed in relative efficacy by imazapicat 0.22 kg a.i. ha–1 plus methylated seed oil at 0.6% vol/vol(97% top kill), trichlopyr at 0.86 kg a.i. ha–1 and 2,4-D at


3.05 kg a.i. ha–1 (80% top kill), and dicamba at 1.14 kg a.i.ha–1 (66% top kill) (Miller and Lucero 1999).

More thorough tests on H. mantegazzianum were con-ducted by Davies and Richards (1985) in the UnitedKingdom, who applied various herbicides to 24-m2 plotsduring May and documented percent mortality, emergenceof new H. mantegazzianum seedlings, and recolonization byother species over the ensuing two growing seasons.Trichlopyr at 1.44 kg a.i. ha–1 was most effective, killing theplants within 1 wk and allowing recolonization of plots witha mixture of other species with few new H. mantegazzianumseedlings. Glyphosate at 2.16 kg a.i. ha–1 resulted in 100%mortality of existing plants over a 10-wk period, but manynew seedlings appeared. The residual herbicides chlorsul-furon + methabenzthiazuron (GLEAN C, 70% w.p.,Dupont) at 2.3 kg a.i. ha–1 and metsulfuron-methyl at 0.016kg a.i. ha–1 provided 100% mortality with no seedling emer-gence, but recolonization by other species was sparse.Fosamine-ammonium at 4.8 kg a.i. ha–1 caused chlorosis butdid not kill or prevent flowering of the plants.

In Ireland, repeated glyphosate treatments on a large pop-ulation over a 4-yr period almost completely eliminated H.mantegazzianum (Caffrey 2001). Glyphosate, trichlopyr, orimazapyr, applied early in the growing season, are recom-mended treatments in the British Isles (Wright 1984;Caffrey 1994; Dodd et al. 1994; Tiley et al. 1996).Glyphosate is widely used on H. mantegazzianum, even incountries where its general use has been restricted(Lundström and Darby 1994; Buttenschøn 2003). A 4-yreradication protocol for H. mantegazzianum populationswithin a single drainage catchment using glyphosate wasoutlined by Caffrey (1999). As with most plants, the effectsof glyphosate may appear slowly, often beginning about amonth after application as a yellowing of newer leaves, fol-lowed by a complete yellowing (Stensones and Garnett1994).

12. Response to Other Human ManipulationsVarious means have been used to control H. mantegazz-ianum including root cutting, mowing, severing of the flow-ering stem or umbels, grazing, trampling and soil tilling.Severing of the roots 8–12 cm below the soil surface is usu-ally very effective in killing the plant (Tiley et al. 1996;Buttenschøn 2003; Nielsen et al. 2005). This should be doneearly in the spring and again in the summer to catch escapesand the plants either removed or left to dry. Hand pulling iseffective only with very young seedlings. Mowing or scyth-ing should be repeated several times during the growing sea-son in order to be effective, but even a single spring cuttingcan reduce the number and size of seeds (Tiley et al. 1996).Care must be taken not to spread seeds during mowing oper-ations. Removal of umbels during early flowering stages canresult in considerable reduction in seed production, but newumbels, usually with fewer flowers and less viable seed, willform on lower branches (Pysek et al. 1998). Removal ofumbels when seeds are formed but have not yet matured isrecommended by Buttenschøn (2003); however, in order toavoid the possibility of seeds maturing on the severed

umbels, they should be removed from the site anddestroyed.

Tiley et al. (1996) noted that tilling is effective for con-trolling infestations on agricultural land. While this may notappear to be applicable under most Canadian conditions,particularly riparian areas, it is a consideration near residen-tial areas. Grazing, especially by sheep, has also been foundto be an effective control (Andersen 1994), although part ofthis effectiveness may be due to trampling (see also Morton1978).

No references regarding the effectiveness of fire or flametreatments have been found. However, heat treatment of H.mantegazzianum using the “Waipuna” system has been sug-gested as an effective control method (Sanh 2003). It hasbeen tried in the Nanaimo–Parksville area of BritishColumbia with varying results (Wall, personal observation).

Biological control agents are also under investigation(Fowler et al. 1991; Seier 2003; Wittenberg et al. 2003;Seier et al. 2004). Indigenous pathogens and foraginginsects of Heracleum spp. have been observed and cata-logued in Europe (Bem and Murant 1979; Sampson 1994;Bürki and Nentwig 1997; Jakob et al. 1998; see Section 13),but possibilities of enhancing the effectiveness of these nat-ural enemies in integrated weed control strategies are onlybeginning to be explored (Erneberg et al. 2003; de Voogd etal. 2003). Erneberg et al. (2003) and de Voogd et al. (2003)have conducted preliminary tests on the fungus Sclerotiniasclerotiorum (Lib.) de Bary as a potential bioherbicide.Seier (2003, 2005) has screened numerous fungi collectedfrom H. mantegazzianum in its native range and found threepromising candidates for biological control: Phloeosporaheraclei (Lib.) Petr., Ramulariopsis sp. and Septoria hera-cleicola Kabat & Bubak.

13. Response to Herbivory, Disease and HigherPlant Parasites(a) Herbivory — Herbivory by mammals, insects and otherorganisms does occur in H. mantegazzianum, but it does notappear to be a limiting factor for populations in Canada. Theproduction of furanocoumarins and lignans in many speciesof Apiaceae, including H. mantegazzianum, confers adegree of protection from herbivory by many insects(Berenbaum 1981; Klocke et al. 1989; Zangerl and Nitao1998; Harmatha and Nawrot 2002). Also, much of the veg-etative growth of H. mantegazzianum occurs in the winterand early spring, especially in maritime climates, whenthere is limited activity by herbivorous species.

(i) Mammals — Some farmers near Parksville, BC haveindicated that their livestock (cattle, goats, hogs) havehelped to suppress H. mantegazzianum in pastures.Browsing injury has been observed in areas frequented bydeer on Vancouver Island, BC (Wall, personal observation).In the United Kingdom, cattle, sheep, goats and hogs areallowed to pasture on H. mantegazzianum but plants thatescape the effects of grazing can flower and produce seed(Wright 1984; Tiley et al. 1996). In an experimental study inDenmark, 7 yr of grazing by 5–10 sheep per hectare eradi-cated H. mantegazzianum (Andersen and Calov 1996).


Sheep grazing is considered the most promising of non-chemical control method on suitable sites with dense popu-lations.

(ii) Birds and/or other vertebrates — Clegg and Grace(1974) report that, in Scotland, seeds left in areas baited forbirds were uneaten. No other information has been found.

(iii) Insects — No published records of specific insects onH. mantegazzianum have been found for Canada, althoughAlex (1992) noted that inflorescences in Ontario were oftenheavily infested with aphids. Plants in the Royal BotanicGarden in Hamilton, ON, were damaged by an aphid species(Hemiptera: Issidae) in 2004 (Hamilton and Rothfels, per-sonal communication). On Vancouver Island, BC, leafminer damage is frequently observed during the summermonths (Wall, personal observation). Young plants cultivat-ed at Ottawa, ON, in 2004 were attacked and heavily dam-aged by larvae of the butterfly Papilio polyxenes Farbicus(Darbyshire, personal observation). The insects Cavariellapastinacae (Linn.) (Hemiptera: Aphididae) and Phytomyzasphondylii Goureau (Diptera: Agromyzidae) have beenstudied on H. maximum in Canada with respect to fura-nocoumarin resistance (Camm et al. 1976b; Ashwood-Smith et al. 1984). Likewise, Berenbaum (1981) studiedfuranocoumarin resistance in several insect species found onH. mantegazzianum in New York State and found thatspecies with resistance to both linear and angular fura-nocoumarins are likely to have a restricted host range, main-ly Heracleum, Angelica and Pastinaca, while there were afew other insects, largely restricted to Apiaciae, that wereresistant only to linear furanocoumarins.

In Europe, many insect species have been recorded on H.mantegazzianum. Jakob et al. (1998) recorded 55 species onH. mantegazzianum in Switzerland, mainly in sunny, low-land sites but none appeared to significantly affect planthealth. Hendrix (1984) found that floral herbivory of H.maximum in Iowa by the parsnip webworm Depressariapastinacella (Duponchel) (Lepidoptera: Oecophoridae)resulted in some losses in number and viability of seed, butmuch of these losses were offset by an increase in seed setby the normally sterile secondary umbels. The followingparagraphs provide a summary of insect taxa recorded fromH. mantegazzianum, mainly from Europe.

Coleoptera: the weevils (Curculionidae) Apion apricansHerbst (Bürki and Nentwig 1997), Apion flavipes (Payk.)(Jakob et al. 1998), Apion gibbirostre (Gyllenhal)(Gassmann and Kok 2002), Barypeithes pellucidusBoheman (Jakob et al. 1998), Euthron fagi Linn. (Bürki andNentwig 1997; Jakob et al. 1998), Liparus germanus (Linn.)(Jakob et al. 1998), Lixus iridis Ol. (Wittenberg et al. 2003),Phyllobius calcaratus Fabricius (Jakob et al. 1998),Liophloeus tessulatus (Müller) (Bürki and Nentwig 1997;Jakob et al. 1998); the scarab beetles (Scarabidae) Hopliafarinosa (Linn.) (Jakob et al. 1998) and Trichius fasciatus(Linn.) (Jakob et al. 1998); the long-jointed beetle(Lagriidae) Lagria hirta (Linn.) (Bürki and Nentwig 1997;Jakob et al. 1998); the long-horned beetles (Cerambycidae)

Leptura rubra (Linn.) (Jakob et al. 1998) and Strangaliamelanura (Linn.) (Jakob et al. 1998); and, the pollen beetle(Nitulidae) Meligethes aeneus (Stephens) (Bürki andNentwig 1997; Jakob et al. 1998).

Diptera: the leaf miners (Agromyzidae) Euleia fratria(Loew) (Berenbaum 1981), Euleia heraclei (Linn.) (Sampson1994), Melanogromyza angeliciphaga (Spencer) (Bürki andNentwig 1997; Jakob et al. 1998), Phytomyza sphondylii(Ashwood-Smith et al. 1984; Sampson 1994, Bürki andNentwig 1997; Jakob et al. 1998), and the carrot fly(Psilidae) Psila rosae (Fabricius) (Hardman and Ellis 1982;Degen et al. 1999).

Hemiptera: the aphids (Aphididae) Anuraphis subterranea(Walker) (Bürki and Nentwig 1997), Aphis fabae Scopoli(Bürki and Nentwig 1997), Cavariella aegopodii (Scopoli)(Sampson 1994), Cavariella pastinacae (Sampson 1994;Bürki and Nentwig 1997), Cavariella theobaldi (Gilletteand Bragg) (Sampson 1994). Paramyzus heraclei Börner(Sampson 1994); the plant bugs (Miridae) Calocoris affinis(Herrich-Schäeffer) (Bürki and Nentwig 1997; Jakob et al.1998), Calocoris sexguttatus (Fabricius) (Bürki andNentwig 1997; Jakob et al. 1998), Dicyphus errans (Wlff.)(Bürki and Nentwig 1997; Jakob et al. 1998), Orthopsbasilis Costa (Sampson 1994), Orthops campestris (Linn.)(Anonymous 2003b; Collins 2003), Orthops kalmii (Linn.)(Bürki and Nentwig 1997; Jakob et al. 1998), Orthopsscutellatus Uhler (Berenbaum 1981); the leafhoppers(Cicadellidae) Eupteryx atropunctata (Goeze) (Jakob et al.1998), Idiocerus populi (Linn.) (Jakob et al. 1998), Eupteryxaurata (Linn.) (Sampson 1994; Bürki and Nentwig 1997;Jakob et al. 1998), Aphrodes bicincta (Schrank) (Bürki andNentwig 1997; Jakob et al. 1998); as well as Graphosomaitalicum (Linn.) (Pentatomidae) (Jakob et al. 1998),Philaenus spumaris (Linn.) (Cercopidae) (Berenbaum 1981;Sampson 1994; Jakob et al. 1998), and Trioza apicalis(Förster) (Psyllidae) (Sampson 1994).

Lepidoptera: the leafrollers (Oecophoridae) Agonopterixclemensella Chambers (Berenbaum 1981), Agonopterix her-acleana (Linn.) (Sampson 1994), and Depressaria pastina-cella (Duponchel) (Berenbaum 1981; Sampson 1994; Bürkiand Nentwig 1997; Jakob et al. 1998); and Cydia aurana(Fabricius) (Tortricidae) (Sampson 1994), Epermeniachaerophyllella (Goeze) (Epermeniidae) (Sampson 1994),and Mamestra brassicae (L.) Barathra (Noctuidae) (Seier etal. 2004).

Thysanoptera: the thrips Thrips atratus Haliday and Thripsvulgatissimus Haliday (Sampson 1994).

(iv) Nematodes and/or other non-vertebrates — No infor-mation has been found on nematodes. Slugs have beenobserved feeding on H. mantegazzianum stems during latesummer on Vancouver Island, BC (Wall, personal observa-tion). In Switzerland, Bürki and Nentwig (1997) noticedslugs and snails causing considerable damage to H. spho-ndylium, but rarely attacking H. mantegazzianum.


(b) Diseases(i) Fungi — No records of fungi on H. mantegazzianum inNorth America have been found, although there are numer-ous records for H. maximum (Savile 1965; Conners 1967;Ginns 1986; Farr et al. 2005). Worldwide, the followingfungi have been documented: Erysiphe heraclei DC.(Ascomycota: Erysiphales) (Farr et al. 2005); Peronosporaheraclei Rabenh. (Oomycota: Peronosporales) (Farr et al.2005); Phloeospora heraclei (Lib.) Petr. (Deuteromycetes:Coelomycetes) (Farr et al. 2005; Seier 2003); Phoma com-planata (Tode: Fr.) Desm. (Deuteromycetes: Coelomycetes)(Sampson 1994; Farr et al. 2005); Phoma sp.(Deuteromycetes: Coelomycetes) (Seier et al. 2004);Ramulariopsis sp. (Deuteromycetes: Hyphomycetes) (Seier2003); Ramularia heraclei (Oudem.) Sacc.(Deuteromycetes: Hyphomycetes) (Seier et al. 2004);Sclerotinia sclerotiorum (Ascomycetes: Helotiales) (Grayand Noble 1965; Farr et al. 2005; Erneberg et al. 2003); andSeptoria heracleicola Kabát & Bubák (Deuteromycetes:Coelomycetes) (Farr et al. 2005; Seier 2003).

The fungi Phloeospora heraclei, Ramulariopsis sp.,Sclerotinia sclerotiorum and Septoria heracleicola areunder investigation as potential biological control agents forH. mantegazzianum (Seier 2003; Erneberg et al. 2003; deVoogd et al. 2003).

(ii) Bacteria — Fischer et al. (1976, 1982) isolatedPseudomonas syringae van Hall from H. mantegazzinumroots with soft rot. Sampson (1994) observed soft rots ofroots and lower stems of both H. mantegazzianum and H.sphondylium in the United Kingdom but did not determinethe causal agents.

(iii) Viruses — No records of virus diseases in H. man-tegazzianum have been found in Canada. Zitter (2001) listsH. mantegazzianum as a host of celery mosaic virus in theUnited States. In the United Kingdom, Sampson (1994)detected aphid-transmitted agents that caused yellow mot-tling but made no identifications. Viruses from symptomlessH. sphondylium have been identified as celery mosaic virus,cow-parsnip mosaic virus, parsnip yellow fleck virus, carrotred leaf virus, carrot mottle virus, heracleum latent virus,anthriscus yellows virus, and heracleum viruses 3–6 (Bemand Murant 1979; Brunt et al. 1996).

(c) Higher Plant Parasites — No records found.

14. PrognosisThe spread of H. mantegazzianum is likely to continuethrough much of southern Canada over the next 25–100 yrwith worsening ecological, economic, and health effects. Itsrange expansion will likely follow a pattern similar to othertemperate regions with initial establishment in urban gar-dens followed by escape and increasingly rapid spread intoriparian areas, forest margins, and unmanaged urban areasin the broader landscape. It is unclear how far north it willspread in Canada and whether it will disperse into naturalareas, particularly forested areas away from human settle-ments. Its occurrence in Kapuskasing, ON, indicates it is tol-

erant of cold, continental climates. Its wide establishmentacross the country and continued sale as an ornamental plantsuggest that even the implementation of a coordinated andcomprehensive management strategy is unlikely to eradicatethe species.

A variety of control measures are available for H. man-tegazzianum with the most efficacious strategy dependingon the habitat and landscape in which the population occurs,as well as the population size and growth stage of the plants(Dodd et al. 1994). Reduction of seed production is animportant part of any control strategy since the plant is aprolific seed producer and the seeds may remain viable inthe soil for many years (Caffrey 1999). Because of its peren-nial habit and the persistent seed bank, eradication of estab-lished H. mantegazzianum populations may require annualtreatments for as long as 10 yr (Andersen and Calov 1996;Buttenschøn 2003). The monocarpic perennial habit of H.mantegazzianum includes several years of vegetativegrowth during which there is opportunity for control.Similarly, the removal of seed heads prevents dispersal andprovides an interim measure during which time more com-prehensive control measures can be developed. Herbicides,animal grazing and mechanical methods are all successful;however, the toxic nature of the sap requires that precau-tions must be taken by those using manual methods. In addi-tion, the prevalence of H. mantegazzianum populationsalong steep embankments, in deep ravines and other inac-cessible places, makes manual treatments difficult. Toreduce damage to surrounding vegetation, herbicide appli-cations, particularly glyphosate, are best done as spot spray-ing or swiping early in the growing season.

A longer-term but widely accepted approach to manage-ment of invasive plants is through the deployment of bio-logical controls (Huffaker and Caltagirone 1986; Bellows2001) and this approach is underway in Europe as part of aninternational effort to manage H. mantegazzianum (Nielsenet al. 2005). Although much of the ground-breakingresearch may take place outside Canada, we have numerousessential roles, including support and cooperation for thisresearch, testing of new control methods, and monitoringpopulations of natural enemies. As with most weeds, suc-cess of biological controls will likely depend on the numberof natural enemies affecting the plant (Memmott et al. 2000;Denoth et al. 2002) and on an integrated approach to control(Henne et al. 2005). There appear to be many possibilitiesfor using both classical and inundative biological controls aspart of integrated management strategies for H. mantegazz-ianum. Some ecological advantages of H. mantegazzianumover other species, such as early spring growth, could pro-vide opportunities for biological controls that exploit coldtolerant agents. The toxic nature of furanocoumarins to mostherbivores also suggests a potential for highly specializedbiological control agents. Biological controls cannot beviewed in isolation or to the exclusion of manual or chemi-cal control methods but need to be integrated into the totalmanagement scenario.

We recommend that three strategies be used to controlrange expansion and population growth of H. mantegazz-ianum in Canada. First, we stress that prevention of intro-


ductions into new regions is more effective than large-scalecontrol programs later in the invasion process. Canada hasan opportunity to learn from the invasion dynamics inEurope and focus management attention on H. mantegazz-ianum before it reaches the exponential growth phase andpopulations begin to spread into the broader landscape.Where most populations are small but widespread in aregion like the Lower Fraser Valley or southeasternVancouver Island, invasion ecology theory recommendsthat control efforts focus on populations on the margin ofrange expansion as the most effective method for slowing orpreventing further invasion (Moody and Mack 1988). Thisstrategy may be particularly effective for controlling H.mantegazzianum in recently invaded regions because it tar-gets small and young populations that may not have reachedflowering age and established seed banks. However, largepopulations should not be neglected because they are theprimary source of seeds for dispersal, particularly alongrivers and streams (Wadsworth et al. 2000). Second, whilethe initial spread of H. mantegazzianum into new sites orregions is likely because of human transport (cf. Case andBeaman 1992; Pysek et al. 1998), subsequent dispersal isprimarily by water or the movement of soil contaminatedwith seed. This suggests that control focusing on large,flowering populations along streams or urban stormwatersystems, as well as managing the movement of soil contam-inated with seed, can also reduce local population spread.Since the primary non-human dispersal mechanism is water,management strategies must be coordinated throughout anentire drainage system if control is to be effective (Tiley andPhilp 1992; Caffrey 2001). Finally, public education is crit-ical for effective management. The close association of H.mantegazzianum with urban and suburban areas in Canadaincreases opportunities for landowners, stewardship groups,and park users to participate in control programs. Publiceducation can be particularly effective in reducing the trans-port of seeds or plants for horticultural plantings, and iden-tifying new populations before they expand. As well,regulation of horticultural propagation and sale should beused to reduce range expansion in Canada.

ACKNOWLEDGEMENTSWe thank the following people who kindly provided distri-bution records or other information for this article: RobAdam (Parksville, BC), Eva Antonijevic (North Vancouver,BC), Wasyl Bakowsky (Ontario Natural Heritage InformationCentre), Adolf Ceska (Victoria, BC), Carol Cornish(Parksville, BC), Bill Crins (Ontario Ministry of NaturalResources), Neil Dawe (Qualicum Beach, BC), GarryFletcher (Victoria, BC), Hugh Griffith (Richmond, BC),Andy Hamilton (Agriculture and Agri-Food Canada), SharonHarcourt (Victoria, BC), Matt Henderson (West Vancouver,BC), Michael Hunter (North Vancouver, BC), Steve Jenkins(West Vancouver, BC), Michael Jessen (Qualicum Beach,BC), Brian Klinkenberg (University of British Columbia),Rose Klinkenberg (Richmond, BC), Frank Lomer(Vancouver, BC), John Morton (University of Waterloo),Mike Oldham (Ontario Natural Heritage Information Centre),Gail Pettinger (Victoria, BC), Adriane Pollard (Saanich, BC),

Dave Polster (Duncan, BC), Hans Roemer (Saanich, BC),Beth Ross (Victoria, BC), Carl Rothfels (Royal BotanicalGardens, Hamilton, ON), Faye Smith (Qualicum Beach,BC), Don Sutherland (Ontario Natural Heritage InformationCentre), Terry Taylor (Vancouver, BC), Rob Walker(Victoria, BC), and Harry Williams (Duncan, BC).

Margo Murray of Agriculture and Agri-Food Canada isthanked for assistance. Barry Flahey of Agriculture andAgri-Food Canada produced the illustrations in Figure 1.We also thank herbarium curators at CAN, DAO, HAM,MICH, MTMG, NSPM, OAC, QUE, TRT, UBC, UNB, andUWO for the loan of specimens for study.

Alex, J. F. 1992. Ontario weeds, descriptions, illustrations, andkeys to their identification. Ontario Ministry of Agriculture andFood. Toronto, ON. 304 pp.Andersen, U. V. 1994. Sheep grazing as a method of controllingHeracleum mantegazzianum. Pages 77–92 in L. C. de Waal, L. E.Child, P. M. Wade, and J. H. Brock, eds. Ecology and managementof invasive riverside plants. John Wiley & Sons, Chichester, UK. Andersen, U. V. and Calov, B. 1996. Long-term effects of sheepgrazing on giant hogweed (Heracleum mantegazzianum).Hydrobiologia 340: 227–284.Andrews, A. H., Giles, C. J. and Thomsett, L. R. 1985. Suspectedpoisoning of a goat by giant hogweed. Vet. Rec. 116: 205–207.Anonymous. 1999. Category DY, Federal Noxious Weed List.USDA–APHIS–PPQ. [Online] Available: http://invader.dbs.umt.edu/Noxious_Weeds/ [2005 Dec. 02].Anonymous. 2002. Photodermatitis from giant hogweed. CDSummary, an epidemiology publication of the Oregon Departmentof Human Services. [Online] Available: http://www.oregon.gov/DHS/ph/cdsummary/2002/ohd5110.pdf [2005 Dec. 02].Anonymous. 2003a. Giant hogweed (Heracleum mantegazz-ianum). Washington State Noxious Weed Control Board. [Online]Available: http://www.nwcb.wa.gov/weed_info/Written_find-ings/Heracleum_mantegazzianum.html [2004 Dec. 02].Anonymous. 2003b. Entomofauna of the Kuril Islands, Chapter 3,Sect. 4. Order Heteroptera. Int. Kuril Project. [Online] Available:http://artedi.fish.washington.edu/okhotskia/ikip/Results/publica-tions/entobook/chapter3–4 [2005 Dec. 02].Anonymous. 2005. The Wildlife and Countryside Act 1981,Schedule 1, Part II. [Online] Available: http://www.naturenet.net/law/wca.html [2005 Dec. 02].Arora, K., Grace, J. and Stewart, F. 1982. Epidermal features ofHeracleum mantegazzianum Somm. & Lev., H. sphondylium L.and their hybrid. Bot. J. Linn. Soc. 85: 169–177.Ashwood–Smith, M. J., Ring, R. A., Liu, M., Phillips, S. andWilson, M. 1984. Furanocoumarin resistance in the larvae ofPhytomyza sphondylii (Diptera: Agrimyzidae) feeding onHeracleum lanatum is associated with the enzymatic breakdown of8–methoxypsoralen. Can. J. Zool. 62: 1971–1976.Barclay, G. F., Oparka, K. J. and Johnson, R. P. C. 1977.Induced disruption of sieve tube element plastids in Heracleummantegazzianum. J. Exp. Bot. 28: 709–717.Baskin, C. C. and Baskin, J. M. 1998. Seeds, ecology, biogeog-raphy, and evolution of dormancy and germination. AcademicPress, San Diego, CA. 666 pp.Bell, C. R. and Constance, L. 1966. Chromosome numbers in theUmbelliferae. III. Am. J. Bot. 53: 512–520.Bellows, T. S. 2001. Restoring population balance through naturalenemy introductions. Biol. Contr. 21: 199–205.Bem, F. and Murant, A. F. 1979. Transmission and differentia-tion of six viruses infecting hogweed (Heracleum sphondylium) inScotland. Ann. Appl. Biol. 92: 237–242.


Berenbaum, M. 1981. Patterns of furanocoumarin distributionand insect herbivory in Umbelliferae: plant chemistry and commu-nity structure. Ecology 62: 1254–1266.Brummitt, R. K. 1968. Heracleum L. Pages 364–366 in T. G.Tutin, V. H. Heywood, N. A. Burges, D. M. Moore, D. H.Valentine, S. M. Walters, and D. A. Webb, eds. Flora Europaea.Vol. 2. Rosaceae to Umbelliferae. Cambridge University Press,Cambridge, UK.Brummitt, R. K. 1971. Relationship of Heracleum lanatumMichx. of North America to H. sphondylium of Europe. Rhodora73: 578–584.Brunt, A. E., Crabtree, K., Dallwitz, M., Gibbs, A., Watson, L.and Zucher, E. 1996. Plant viruses online. [Online] Available:http://image.fs.uidaho.edu/vide/refs.htm [2005 Dec. 02].Bürki, C. and Nentwig, W. 1997. Comparison of herbivore insectcommunities of Heracleum sphondylium and H. mantegazzianumin Switzerland (Spermatophora: Apiaceae). Entomol. Gen. 22:147–155.Buttenschøn, R. M. 2003. Experiences from management of gianthogweed (Heracleum mantegazzianum) in Denmark and resultsfrom a pilot study on nonchemical control methods: abstract.Botanical Electronic News [Online] Available: www.ou.edu/cas/botany-micro/ben/ben316.html [2005 Dec. 02].Caffrey, J. M. 1994. Spread and management of Heracleum man-tegazzianum (Giant Hogweed) along Irish river corridors. Pages67–76 in L. C. de Waal, L. E. Child, P. M. Wade, and J. H. Brock,eds. Ecology and management of invasive riverside plants. JohnWiley & Sons, Chichester, UK.Caffrey, J. M. 1999. Phenology and long-term control ofHeracleum mantegazzianum. Hydrobiologia 415: 223–228.Caffrey, J. M. 2001. The management of giant hogweed in anIrish river catchment. J. Aquat. Plant Manag. 39: 28–33.Camm, E., Buck, H. W. L. and Mitchell, J. C. 1976a.Phytodermatitis from Heracleum mantegazzianum. ContactDermatitis 2: 68–72.Camm, E. L., Wat, C. K. and Towers, G. H. N. 1976b.Assessment of the roles of furanocoumarins in Heracleum lana-tum. Can. J. Bot. 54: 2562–2566.Carbonnier, J., Fatianoff, O. and Molho, D. 1982. Phytochimiecompare des taxons rattaches à la tribu des Peucedaneae. Pages387–513 in A.-M. Cauwet-Marc and J. Carbonnier, eds. Actes du2ème symposium international sur les Ombellifères (Perpignan1977). Contributions pluridisciplinaires à la Systématique.Missouri Botanical Garden Monographs in Systematic Botany 6,St. Louis, MO.Case, M. A. and Beaman, J. H. 1992. Heracleum mantegazz-ianum (giant cow parsnip): another exotic in the Michigan Flora.Mich. Bot. 31: 152–154.Cerceau-Larrival, M.-Th. 1971. Morphologie pollinique et cor-rélations phylogénétiques chez les Ombellifères. J. Linn. Soc. 64(Suppl. 1): 109–155.Clarke, C. H. 1974. Giant hogweed sap: another environmentalmutagen. Mutat. Res. 31: 63.Clegg, L. M. and Grace, J. 1974. The distribution of Heracleummantegazzianum (Somm. & Levier) near Edinburgh. Trans. Proc.Bot. Soc. Edinb. 42: 223–229.Collingham, Y. C., Wadsworth, R. A., Huntley, B. and Hulme,P. E. 2000. Predicting the spatial distribution of non-indigenousriparian weeds: issues of spatial scale and extent. J. Appl. Ecol. 37:13–27.Collins, A. 2003. Recent reports. Fauna reports. This is Hampshire.Southampton Nat. Hist. Soc. Page 8. [Online] Available:http://www.communigate.co.uk/hants/snhs/page5.phtml [2005 Dec.02].

Conners, I. L. 1967. An annotated index of plant diseases inCanada and fungi recorded on plants in Alaska, Canada andGreenland. Canada Department of Agriculture, Ottawa, ON. Publ.1251. 381 pp.Cronquist, A., Holmgren, N. H. and Holmgren, P. K. 1997.Intermountain flora. Vascular plants of the Intermountain West,U.S.A. Volume 3, Part A. Subclass Rosidae (except Fabales). TheNew York Botanical Garden, Bronx, NY. 446 pp.Darbyshire, S. J., Favreau, M. and Murray, M. 2000. Commonand scientific names of weeds in Canada/Noms populaires et sci-entific des plantes nuisibles du Canada. Research Branch,Agriculture and Agri-Food Canada/Direction général de larecherche, Agriculture et Agroalimentaire Canada, Ottawa, ON.Publ. 1397/B. 132 pp.Darbyshire, S. J. 2003. Inventory of Canadian AgriculturalWeeds. Agriculture and Agri-food Canada, Research Branch,Ottawa, ON. 396 pp. [Online] Available: http://res2.agr.ca/ecorc/weeds_herbes/title-titre_e.htm [2005 Dec. 02].Davies, D. H. K. and Richards, M. C. 1985. Evaluation of herbi-cides for control of giant hogweed (Heracleum mantegazzianumSomm. & Lev.) and vegetation regrowth in treated areas. Tests ofAgrochemicals and Cultivars. Ann. Appl. Biol. 106: 100–101.Dawe, N. K. and White, E. R. 1979. Giant cow parsnip(Heracleum mantegazzianum) on Vancouver Island, BritishColumbia. Can. Field-Nat. 93: 82–83.Dawson, F. H. and Holland, D. 1999. The distribution in banksidehabitats of three alien invasive plants in the UK in relation to the devel-opment of control strategies. Hydrobiologia 415: 193–201.de Voogd, W. B., de Jong, M. D. and Nielsen, C. 2003. Use ofSclerotinia sclerotiorum as a mycoherbicide to control the spreadof giant hogweed (Heracleum mantegazzianum) in the Netherlandsand Denmark. Biological Electronic News 318: 1–3. [Online]Available: http://www.ou.edu/cas/botany-micro/ben/ben318.html[2005 Dec. 02].Degen, T., Staedler, E. and Ellis, P. R. 1999. Host-plant suscep-tibility to the carrot fly, Psila rosae: suitability of various hostspecies for larval development. Ann. Appl. Biol. 134: 27–34.Denoth, M., Frid, L. and Myers, J. H. 2002. Multiple agents inbiological control: improving the odds. Biol. Contr. 24: 20–30.Devagiri, G. M., Singh, J. M., Chand, R. and Srivastava, L. J.1997. Genetic variability in Heracleum candicans Wall. Indian J.Genet. Plant Breed. 57: 280–286.Dickson, J. H. 2001. Alien vascular plants in Scotland: conceptsand consequences – Scotland no Hawai’i. Glasgow Naturalist 23(Suppl.): 2–12.District of Saanich. 2005. Noxious Weeds Bylaw, 2000, No. 8080[Amendment to Bylaw No. 8080 (Noxious Weed)]. 2005 Jul. 13.Dodd, F. S., de Waal, L. C., Wade, P. M. and Tiley, G. E. D.1994. Control and management of Heracleum mantegazzianum(Giant Hogweed). Pages 111–126 in L. C. de Waal, L. E. Child, P.M. Wade, and J. H. Brock, eds. Ecology and management of inva-sive riverside plants. John Wiley & Sons, Chichester, UK.Dragan, T., Zgórka, G. and G`owniak, K. 1999. The content ofsome linear furanocoumarins in the fruits of plants from Heracleumsp. Herba Polonica 45: 12–17 [In Polish, with English abstract].Drever, J. C. and Hunter, J. A. A. 1970. Giant hogweed der-matitis. Scott. Med. J. 15: 315.Dutton, I. 2004. A monster of pain, hogweed stalks capital withtoxic ooze. Times-Colonist, Victoria, BC. 2004 Jul. 24.Epstein, W. L. 1987. Plant-induced dermatitis. Ann. Emerg. Med.16: 950–955.Erneberg, D., Strandberg, B. and Jensen, B. D. 2003.Susceptiblity of a plant invader to a pathogenic fungus: an experi-mental study of Heracleum mantegazzianum (Giant Hogweed) and


Sclerotinia sclerotiorum. Pages 355–372 in L. E. Child, J. H.Brock, G. Brundu, K. Prach, P. Pysek, P. M. Wade, and M.Williamson, eds. Plant invasions: Ecological threats and manage-ment solutions. Bachhuys Publishing, Leiden, the Netherlands.Ezeala, D. O., Hart, J. W. and Sabnis, D. D. 1974. Fractionationof monovalent ion simulated nucleotide triphosphatase activity inextracts of petiolar tissues. J. Exp. Bot. 25: 1037–1044.Farr, D. F., Rossman, A. Y., Palm, M. E. and McCray, E. B.2005. Fungal Databases, Systematic Botany and MycologyLaboratory, ARS, USDA. [Online] Available: http://nt.ars–grin.gov/fungaldatabases/ [2005 Dec. 02].Fensom, D. S., Chattenburg, R., Chung, T., Lee, D. R. andArnold, D. C. 1968. Moving particles in intact sieve tubes ofHeracleum mantegazzianum. Nature (Lond.) 219: 531–532.Fischer, F., van Doorne, H., Lim, H. and Baerheim, A. 1976.Bacteriostatic activity of some coumarin derivatives.Phytochemistry (Oxf.) 15: 1078–1079.Fischer, F., van Doorne, H. and Dannenberg, G. 1982.Glycosides and glycosidase in Heracleum mantegazzianum. Theirpossible role in resistance against fungi. Pages 783–792 in A.-M.Cauwet-Marc and J. Carbonnier, eds. Actes du 2ème symposiuminternational sur les Ombellifères (Perpignan 1977). Contributionspluridisciplinaires à la Systématique. Missouri Botanical GardenMonographs in Systematic Botany 6, St. Louis, MO.Fowler, S. V., Holden, A. N. G. and Schroeder, D. 1991. Thepossibilities for classical biological control of weeds of industrialland in the U.K. using introduced insect herbivores and plantpathogens. Brighton Crop Protection Conference, Weeds 1991. pp.1173–1180.Gagnidze, R. J. and Chkheidze, P. B. 1975. The chromosomenumbers of some species of Greater Caucasus flora. Soobshch.Akad. Nauk Gruz. SSR 79: 433–435. [In Russian and Latin, withEnglish abstract].Gassmann, A. and Kok, L. T. 2002. Musk thistle (nodding this-tle). Pages 229–245 in R. Van Driesche, S. Lyon, B. Blossey, M.Hoddle, and R. Reardon, eds. Biological control of invasive plantsin the eastern United States, US Dep. Agric. For. Serv. Publ.FHTET–2002–04. [Online] Available: http://www.invasive.org/eastern/biocontro/18MuskThistle.html [2005 Dec. 02].Ginns, J. H. 1986. Compendium of plant disease and decay fungiin Canada 1960–1980. Agriculture Canada Publication 1813.Canadian Government Publishing Centre, Ottawa, ON. 416 pp.G`owniak, K., Mroczek, T., Zabza, A. and Cierpic, T. 2003.Isolation and structure elucidation of 5′7–disubstituteds–coumarins in the fruits of Heracleum mantegazzianum.Pharmaceut. Biol. 38: 308–312.Godefroid, S. 1998. Contribution à la connaissance de la distribu-tion d’Heracleum mantegazzianum à Bruxelles. Dumortiera 72:1–7. [In French, with English summary].Grace, J. and Nelson, M. 1981. Insects and their pollen loads at ahybrid Heracleum site. New Phytol. 87: 413–423.Grace, J. and Stewart, F. 1982. Hybridization in the genusHeracleum in the British Isles. Pages 773–782 in A.-M. Cauwet-Marcand J. Carbonnier, eds. Actes du 2ème symposium international sur lesOmbellifères (Perpignan 1997). Contributions pluridisciplinaires à laSystematique. Missouri Botanical Garden Monographs in SystématicBotany 6, St. Louis, MO.Gray, E. G. and Noble, M. 1965. Sclerotinia Disease. Scott.Agric. 44: 265–267.Grime, J. P., Mason, G., Curtis, A. V., Rodman, J., Band, S. R.and Mowforth, M. A. 1981. A comparative study of germinationcharacteristics in a local flora. J. Ecol. 69: 1017–1059.Hallam, J. and Dodd, F. S. 1996. Heracleum mantegazzianum(giant hogweed), how long are the seeds viable? Botanical Societyof the British Isles, News 72: 62.

Hamel, K. S. and Parsons, J. K. 2001. Washington’s aquaticplant quarantine. J. Aquat. Plant Manag. 39: 72–75.Harborne, J. B. 1971. Flavonoid and phenylpropanoid patterns inthe Umbelliferae. Bot. J. Linn. Soc. 64 (Suppl. 1): 293–314.Hardman, J. A. and Ellis, P. R. 1982. An investigation of the hostrange of the carrot fly, Psila rosae. Ann. Appl. Biol. 100: 1–9.Harmatha, J. and Nawrot, J. 2002. Insect feeding deterrent activ-ity of lignans and related phenylpropanoids with a methylene-dioxyphenyl (piperonyl) structure moiety. Entomol. Exp. Appl.104: 51–60.Hart, J. W. and Sabnis, D. D. 1975. The occurrence and someproperties of a new filamentous component of plant extracts. Ann.Bot. 39: 153–159.Harwood, D. G. 1985. Giant hogweed and ducklings. Vet. Rec.116: 300.Hendrix, S. D. 1984. Reactions of Heracleum lanatum to floralherbivory by Depressaria pastinacella. Ecology 65: 191–197.Henne, D. C., Lindgren, C. J., Gabor, T. S., Murkin, H. R. andRoughley, R. E. 2005. An integrated management strategy for thecontrol of purple loosestrife Lythrum salicaria L. (Lythraceae) inthe Netley-Libau Marsh, southern Manitoba. Biol. Contr. 32:319–325.Hindáková, M. and Schwarzová, T. 1987. Karyological study of theSlovak flora XX. Acta Fac. Rerum Nat. Univ. Comen. Bot. 34: 21–25.Hinds, H. R. 2000. The flora of New Brunswick. 2nd ed.University of New Brunswick, Fredericton, NB. 699 pp.Hinterman, J. 1962. Dermatosis of a dog due to Heracleum man-tegazzianum. Schweiz. Arch. Tierheilkd. 109: 654.Holmgren, P. K., Holmgren, N. H. and Barnett, L. C. 1990.Index herbariorium. Part I: The herbaria of the world. 8th ed. NewYork Botanical Garden, Bronx, NY. x + 693 pp.Huffaker, C. B. and Caltagirone, L. E. 1986. The impact of bio-logical control on the development of the Pacific. Agric. Ecosyst.Environ. 15: 95–107.Huxley, A. (ed.) 1992. The new RHS dictionary of gardening. 4Volumes. MacMillan Press, London, UK. 3336 pp.Igali, S., Bridges, B. A., Ashwood–Smith, M. J. and Scott, B. R.1970. Mutagenesis in Escherichia coli, IV. Photosensitization tonear ultraviolet light by 8–methoaxypsoralen. Mutat. Res. 9: 21.Jain, S. R. 1969. Investigations on the essential oil of Heracleummantegassianum L. Planta Med. 17: 230–235.Jakob, B., von Mandach, T. and Nentwig, W. 1998. Phytophagean Heracleum mantegazzianum und Heracleum sphondylium inder Schweiz. Z. Pflanzenkr. Pflanzenschutz. 16: 223-230.Jones, J. G. and Russell, D. G. 1968. Giant hogweed dermatitis.Practitioner 200: 704.Junttila, O. 1975. Allelopathy in Heracleum laciniatum: inhibi-tion of lettuce seed germination and root growth. Physiol. Plant.33: 22–27.Junttila, O. 1976. Allelopathic inhibitors in seeds of Heracleumlaciniatum. Physiol. Plant. 36: 374–378.Kamermans, J. K. 1977. A hairy horror in my Huron haunts.Wood Duck 31: 49–50.Kartesz, J. T. 1999. A synonymized checklist and atlas with bio-logical attributes for the vascular flora of the United States,Canada, and Greenland. 1st ed. in J. T. Kartesz and C. A.Meacham, eds. Synthesis of the North American flora. Version 1.0.North Carolina Botanical Garden, Chapel Hill, NC. [CDROM].Kerr Wood Leidal Associates Ltd. 2005. City of North VancouverEnvironmental Monitoring Program. Burnaby, BC. 65 pp.Kljuykov, E. V., Liu, M., Ostroumova, T. A., Pimenov, M. G.,Tilney, P. M. and van Wyk, B-E. 2004. Towards a standardizedterminology for taxonomically important morphological charactersin the Umbelliferae. S. Afr. J. Bot. 70: 488–496.


Klocke, J. A., Balandrin, M. F., Barnby, M. A. and Yamasaki,R. B. 1989. Limonoids, phenolics, and furanocoumarins as insectantifeedants, repellents, and growth inhibitory compounds. Pages136–149 in J. T. Arnason, B. J. R. Philogene, and P. Morand, eds.Insecticides of plant origin. American Chemical SocietySymposium Series 387. American Chemical Society, Washington,DC.Knapp, H. D. and Hacker, E. 1984. Zur einbürgerung vonTelekia speciosa (Schreb.) Baumg. in Mecklenburg. Gleditschia12: 85–106. [In German, with English abstract]Knudsen, E. A. 1983. Seasonal variations in the content of photo-toxic compounds in giant hogweed. Contact Dermatitis 9:281–284.Knuth, P. 1908. Handbook of flower pollination. Volume 2.Translation by J. R. A. Davis. Clarendon Press, Oxford, UK. 703pp.Krinke L., Moravcová L., Pysek P., Jarosík V., Pergl, J. andPerglová, I. 2005. Seed bank in an invasive alien Heracleum man-tegazzianum and its seasonal dynamics. Seed Sci. Res. 15: 239-248.Kühn, K.-D., Weber, H.-Chr., Dehne, H.-W. and Gworgwor,N. A. 1991. Distribution of vesicular-arbuscular mycorrhizal fungion a fallow agricultural site. I. Dry habitat. Angew. Bot. 65:169–185.Kuske, H. 1940. Percutaneous photosensitization by plant juices.Dermatologica (Basel) 82: 273.Lee, D. R., Arnold, D. C. and Fensom, D. S. 1971. Some micro-scopical observations of functioning sieve tubes of Heracleumusing Nomarski optics. J. Exp. Bot. 22: 25–38.Leslie, P. 2003. Keep your eyes open for this nasty weed.Parksville-Qualicum News, Parksville, BC. 2003 Jul. 04.Lundström, H. 1984. Giant hogweed, Heracleum mantegazz-ianum, a threat to the Swedish countryside. Weeds and WeedControl. 25th Swedish Weed Conference 1: 191–200.Lundström, H. 1990. New experiences of the fight against thegiant hogweed, Heracleum mantegazzianum. Weeds and WeedControl. 30th Swedish Weed Conference 2: 51–58.Lundström, H. and Darby, E. 1994. The Heracleum mantegazz-ianum (Giant Hogweed) problem in Sweden: suggestions for man-agement and control. Pages 93–100 in L. C. de Waal, L. E. Child,P. M. Wade, and J. H. Brock, eds. Ecology and management ofinvasive riverside plants. John Wiley & Sons, Chichester, UK.Mandenova, I. P. 1951. Genus 1069. Heracleum L. Pages161–185 in B. K. Shishkin, ed. Flora of the U.S.S.R. VolumeXVII. Umbelliflorae. [Translation by R. Lavoott, 1974, IsraelProgram for Scientific Translations, Jerusalem, for the NationalScience Foundation and The Smithsonian Institution, Washington,DC.]Marie-Victorin, Frère. 1997. Flore Laurentienne. Troisieme édi-tion mise jour et annotée par L. Brouillet, S. G. Hay, I. Goulet, M.Blondeau, J. Cayouette et J. Labrecque. Les Presses de l’Universitéde Montréal, Montréal, QC. 1093 pp.McIlveen, W. D. 2001. Dispersal of giant cow-parsnip(Heracleum mantegazzianum) along streams in Halton County.Field Botanists of Ontario Newsletter 13 (4): 10–12.Memmott, J., Fowler, S. V., Paynter, Q., Sheppard, A. W. andSyrett, P. 2000. The invertebrate fauna on broom, Cytisus scopar-ius, in two native and two exotic habitats. Acta Oecol. 21:213–222.Merilees, B. 1978. Giant cow parsnip – French Creek. Federationof British Columbia Naturalists Newsletter 16 (4). 1 pp.Merilees, W. 1981. The giant cow parsnip in Greater Vancouver.Discovery (Vancouver Natural History Society) 10: 36–38.

Miescher, G. and Burckhardt, W. 1937. Herakleum dermatitis:case presentation, Swiss Dermatologic Association, Oct. 3–4,1936. Schweiz. Med. Wochenschr. 67: 82. [In German].Miller, T. W. and Lucero, C. 1999. Giant hogweed injury by sev-eral herbicides. [Online] Available: http://mtvernon.wsu.edu/weedscience/publications/hog99.pdf [2005 Dec. 02].Mitchell, J. and Rock, A. 1978. Botanical dermatology.Greengrass Ltd., Vancouver, BC. xiii + 787 pp.Molho, D., Jössang, P., Jarreau, M.-C. and Carbonnier, J.1971. Dérivés furannocoumariniques du genre Heracleum et plusspécialments de Heracleum sprengelianum Wight & Arn. etHeracleum ceylanicum Gardn. ex C. B. Clarke. Étude phy-logénique. J. Linn. Soc. 64 (Suppl. 1): 199–214.Mondoor, A. 2001. Le Grand Livres des Vivaces: Guide Pratique.Les Éditions de l’Homme, Montreal, QC. 391 pp.Moody, M. E. and Mack, R. N. 1988. Controlling the spread of plantinvasions: the importance of nascent foci. J. Appl. Ecol. 25:1009–1021.Moravcová, L., Perglová, I., Pysek, P., Jarosík, V. and Pergl, J.2005. Effects of fruit position on fruit mass and seed germinationin the alien species Heracleum mantegazzianum (Apiaceae) andthe implications for its invasion. Acta Oecol. 28: 1–10.Morton, J. K. 1975. The giant cow parsnip, Heracleum man-tegazzianum, Umbelliferae, in Canada. Can. Field-Nat. 89:183–184.Morton, J. K. 1978. Distribution of giant cow parsnip (Heracleummantegazzianum) in Canada. Can. Field-Nat. 92: 182–185.Murphy, R. 1986. A reanalysis of particle motion in sieve tubes ofHeracleum. Ann. Bot. 57: 667–674.Murray, R. D. H. 2002. The naturally occurring coumarins. W.Herz, H. Falk, G. W. Kirby and R. E. Moore, eds. Progress inchemistry of organic natural products. Springer-Verlag, Vienna,Austria. vii + 673 pp.Nielsen B. E. 1970. Coumarins of umbelliferous plants. Dan.Tidsskr. Farm. 44 (6): 111–286.Nielsen, C., Ravn, H. P., Nentwig, W. and Wade, M. (eds.)2005. The giant hogweed best practices manual: guidelines for themanagement and control of an invasive weed in Europe. Forest andLandscape Denmark, Horsholm. 44 pp. Ochsmann, J. J. 1996. Heracleum mantegazzianum Sommier &Levier (Apiaceae) in Deutschland – Untersuchungen zur Biologie,Verbreitung, Morphologie und Taxonomie. Feddes Repert. 107:557–595. [In German, with English abstract].Otte, A. and Franke, R. 1998. The ecology of the Caucasianherbaceous perennial Heracleum mantegazzianum Somm. et Lev.(Giant Hogweed) in cultural ecosystems of Central Europe.Phytocoenologia 28: 205–232.Page, N. and Wall, R. 2003. Heracleum mantegazzianum: a nastyinvasive plant species in British Columbia. Botanical ElectronicNewsletter 314. [Online] Available: http://www.ou.edu/cas/botany-micro/ben/ben314.html [2005 Dec. 02].Pathak, M. A., Worden, L. R. and Kaufman, K. D. 1967. Effectof structural alterations on the photosensitizing potency of fura-nocoumarins (psoralens) and related compounds. J. Investig.Dermatol. 48: 103.Pira, E., Romano, C., Sulotto, F., Pavan, I. and Monaco, E.1989. Heracleum mantegazzianum growth phases and fura-nocoumarin content. Contact Dermatitis 21: 300–303.Pysek, P. 1991. Heracleum mantegazzianum in the CzechRepublic: dynamics of spreading from the historical perspective.Folia Geobot. Phytotaxon. 26: 439–454.Pysek, P. 1994. Ecological aspects of invasion by Heracleum man-tegazzianum in the Czech Republic. Pages 45–54 in L. C. de Waal,L. E. Child, P. M. Wade, and J. H. Brock, eds. Ecology and man-


agement of invasive riverside plants. John Wiley & Sons,Chichester, UK.Pysek, P. and Prach, K. 1994. How important are rivers for sup-porting plant invasions? Pages 19–26 in L. C. de Waal, L. E. Child,P. M. Wade, and J. H. Brock, eds. Ecology and management ofinvasive riverside plants. John Wiley & Sons, Chichester UK.Pysek, P. and Pysek, A. 1995. Invasion by Heracleum man-tegazzianum in different habitats in the Czech Republic. J. Veg.Sci. 6: 711–718.Pysek, P., Kopecky, M., Jarosík, V. and Kotková, P. 1998. Therole of human density and climate in the spread of Heracleum man-tegazzianum in the Central European landscape. Divers. Distrib. 4:9–16.Roland, A. E. and Zinck, M. 1998. Roland’s Flora of NovaScotia. 3rd ed. 2 Volumes. Nova Scotia Museum & NimbusPublishing, Halifax, NS. 1297 pp.Sampson, C. 1994. Cost and impact of current control methodsused against Heracleum mantegazzianum (giant hogweed) and thecase for instigating a biological control program. Pages 55–66 in L.C. de Waal, L. E. Child, P. M. Wade, and J. H. Brock, eds. Ecologyand management of invasive riverside plants. John Wiley & Sons,Chichester, UK.Sanh, M. 2003. Waipuna and giant hogweed (Heracleum man-tegazzianum). Botanical Electronic Newsletter 316. [Online]Available: http://www.ou.edu/cas/botany-micro/ben/ben316.html[2005 Dec. 02].Savile, D. B. O. 1965. Some fungal parasites of Umbelliferae. Can.J. Bot. 43: 571–596.Seier, M. 2003. Classical biological control of giant hogweed(Heracleum mantegazzianum). International Bioherbicide GroupNewsletter, June, 2003. [Online] Available: http://ibg.ba.cnr.it/[2005 Dec. 02].Seier, M. K. 2005. Fungal pathogens as classical biological con-trol agents for invasive weeds – is it a viable concept for Europe.Neobiota 6: 165–176.Seier, M. K., Wittenberg, R., Ellison, C. A., Djeddour, D. H.and Evans, H. C. 2004. Surveys for natural enemies of giant hog-weed (Heracleum mantegazzianum) in the Caucasus region andassessment for their classical biological control potential inEurope. Pages 149–154 in J. M. Cullen, D. T. Briese, D. J.Kriticos, W. M. Lonsdale, L. Morin, and J. K. Scott, eds.Proceedings of the XI International Symposium on BiologicalControl of Weeds, Canberra, Australia, pp. 149–154, CSIROEntomology, Canberra, Australia.Smellie, J.H. 1968. Giant hogweed. Br. Med. J. 3: 123.Sommer, R. G. and Jillson, O. F. 1967. Phytophotodermatitis(solar dermatitis from plants). Gas plant and the wild parsnip. N.Engl. J. Med. 276: 1484–1486.Spanner, D. C. 1970. The electro–osmotic theory of phloem trans-port in light of recent measurements on Heracleum phloem. J. Exp.Bot. 21: 325–334.Stensones, A. and Garnett, R. P. 1994. Controlling invasiveweeds using glyphosate. Pages 183–188 in L. C. deWaal, L. E.Child, P. M. Wade, and J. H. Brock, eds. Ecology and managementof invasive riverside plants. John Wiley & Sons, Chichester, UK.Stewart, F. and Grace, J. 1984. An experimental study ofhybridization between Heracleum mantegazzianum Somm. & Lev.and Heracleum sphondyllium L. ssp. sphondyllium (Umbelliferae).Watsonia 15: 73–83.Stokes, P. 1952. A physiological study of embryo development inHeracleum sphondylium L. I. The effect of temperature on embryodevelopment. Ann. Bot. (Lond.) 16: 441–447.Tiley, G. E. D. and Philp, B. 1992. Strategy for the control ofgiant hogweed (Heracleum mantegazzianum) on the River Ayr in

Scotland. Vegetation Management in Forestry, Amenity andConservation Areas. Asp. Appl. Biol. 29: 463–466.Tiley, G. E. D. and Philp, B. 1994. Heracleum mantegazzianum(giant hogweed) and its control in Scotland. Pages 101–109 in L.C. de Waal, L. E. Child, P. M. Wade, and J. H. Brock, eds. Ecologyand management of invasive riverside plants. John Wiley & Sons,Chichester, UK.Tiley, G. E. D., Dodd, F. S. and Wade, P. M. 1996. Biologicalflora of the British Isles. 190. Heracleum mantegazzianumSommier et Levier. J. Ecol. 84: 297–319.Townsend, M. E., Wright, H. M. and Hopwood, D. A. 1971.Efficient mutagenesis by near ultraviolet light in the presence of 8-methoxypsoralen in Streptomyces. J. Appl. Bacteriol. 74: 799.Tutin, T. G. 1980. Umbellifers of the British Isles. BotanicalSociety of the British Isles Handbook 2, London, UK. 197 pp.Tyree, M. T. and Fensom, D. S. 1970. Some experimental andtheoretical observations concerning mass flow in the vascular bun-dles of Heracleum. J. Exp. Bot. 21: 304–324.Wade, M. 1997. Predicting plant invasions: making a start. Pages1–18 in J. H. Brock, M. Wade, P. Pysek, and D. Green, eds. Plantinvasions: studies from North America and Europe. Bachhuys,Leiden, the Netherlands.Wadsworth, R. A., Collingham, Y. C., Willis, S. G., Huntley, B.and Hulme, P. E. 2000. Simulating the spread and management ofalien riparian weeds: are they out of control? J. Appl. Ecol. 37(Suppl. 1): 28–38.Walker, N. F., Hulme, P. E. and Hoelzel, A. R. 2003. Populationgenetics of an invasive species, Heracleum mantegazzianum:implications for the role of life history, demographics and inde-pendent introductions. Mol. Ecol. 12: 1747–1756.Wanscher, J. H. 1932. Studies on the chromosome numbers of theUmbelliferae. Bot. Tidsskr. 42: 49–59.Webb, C. J., Sykes, W. R. and Garnock-Jones, P. J. 1988. Floraof New Zealand. Volume IV. Naturalized pteridophytes, gym-nosperms, dicotyledons. Botany Division, DSIR, Christchurch,New Zealand. 1365 pp.Weimarck, G., Stewart, F. and Grace, J. 1979. Morphometricand chromatographic variation and male meiosis in the hybridHeracleum mantegazzianum x H. sphondylium (Apiaceae) and itsparents. Hereditas (Lund) 91: 117–127.Westbrooks, R. G. 1991. Fact Sheet, Heracleum mantegazzianumSommier & Levier – Giant hogweed, golpar. Federal NoxiousWeed Inspection Guide. Whiteville Plant Methods Center,Whiteville, NC.Williamson, J. A. and Forbes, J. C. 1982. Giant hogweed(Heracleum mantegazzianum): its spread and control withglyphosate in amenity areas. Proceedings 1982 British CropProtection Conference 3: 967–972.Willis, S. G. and Hulme, P. E. 2002. Does temperature limit theinvasion of Impatiens glandulifera and Heracleum mantegazz-ianum in the UK? Funct. Ecol. 16: 530–539.Wittenberg, R., Poll, M. and Cock, M. 2003. Giant Hogweed(Heracleum mantegazzianum). Quarterly Report, CABIBioscience Switzerland Centre, Delemont, Switzerland, 2003 Jun.30, pp. 17–18.Wright, M. 1984. Giant hogweed: time for action is now (Heracleummantegazzianum, Great Britain). New Sci. 101(1404): 44. Zangerl, A. R. and Nitao, J. K. 1998. Optimal defence, kin con-flict and the distribution of furanocoumarins among offspring ofwild parsnip. Evol. Ecol. 12: 443–457.Zitter, T. A. 2001. A checklist of major weeds and crops as naturalhosts for plant viruses in the Northeast. Cornell University VegetableMD Online. [Online] Available: http://vegetablemdonline.ppath.cor-nell.edu/Tables/WeedHostTable.html [2005 Dec. 02].

Raincoast Applied Ecology - CANADIAN JOURNAL …...The Biology of Invasive Alien Plants in Canada....

Documents

Transcript of Raincoast Applied Ecology - CANADIAN JOURNAL …...The Biology of Invasive Alien Plants in Canada....